mirror of
https://github.com/pgvector/pgvector.git
synced 2026-07-22 20:15:46 +08:00
Compare commits
23 Commits
ivfflat-in
...
fanalyzer
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
ba16f5e7cf | ||
|
|
4a1a91abf9 | ||
|
|
8d7abb6590 | ||
|
|
b247b688a8 | ||
|
|
9672446a4c | ||
|
|
334614b7f7 | ||
|
|
643eacd9dc | ||
|
|
bca50a03fa | ||
|
|
d87833cacc | ||
|
|
4c01073ac1 | ||
|
|
6fed8f1e78 | ||
|
|
611f5b1047 | ||
|
|
e367155732 | ||
|
|
466c556b1a | ||
|
|
75e09265d6 | ||
|
|
77c5070fb6 | ||
|
|
1006fdf3f2 | ||
|
|
4209c9b3af | ||
|
|
ec0bb4e4ff | ||
|
|
b164833933 | ||
|
|
30fb4dd602 | ||
|
|
d032726976 | ||
|
|
8fa9001474 |
4
.github/workflows/build.yml
vendored
4
.github/workflows/build.yml
vendored
@@ -39,6 +39,10 @@ jobs:
|
||||
sudo apt-get update
|
||||
sudo apt-get install libipc-run-perl
|
||||
- run: make prove_installcheck
|
||||
- if: ${{ matrix.os == 'ubuntu-22.04' }}
|
||||
run: make clean && make
|
||||
env:
|
||||
PG_CFLAGS: -Werror -fanalyzer
|
||||
mac:
|
||||
runs-on: macos-latest
|
||||
if: ${{ !startsWith(github.ref_name, 'windows') }}
|
||||
|
||||
2
Makefile
2
Makefile
@@ -8,7 +8,7 @@ HEADERS = src/vector.h
|
||||
|
||||
TESTS = $(wildcard test/sql/*.sql)
|
||||
REGRESS = $(patsubst test/sql/%.sql,%,$(TESTS))
|
||||
REGRESS_OPTS = --inputdir=test --load-extension=vector
|
||||
REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION)
|
||||
|
||||
OPTFLAGS = -march=native
|
||||
|
||||
|
||||
@@ -5,7 +5,7 @@ OBJS = src\hnsw.obj src\hnswbuild.obj src\hnswinsert.obj src\hnswscan.obj src\hn
|
||||
HEADERS = src\vector.h
|
||||
|
||||
REGRESS = btree cast copy functions input ivfflat_cosine ivfflat_ip ivfflat_l2 ivfflat_options ivfflat_unlogged
|
||||
REGRESS_OPTS = --inputdir=test --load-extension=vector
|
||||
REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION)
|
||||
|
||||
# For /arch flags
|
||||
# https://learn.microsoft.com/en-us/cpp/build/reference/arch-minimum-cpu-architecture
|
||||
|
||||
@@ -218,7 +218,6 @@ typedef HnswNeighborTupleData * HnswNeighborTuple;
|
||||
typedef struct HnswScanOpaqueData
|
||||
{
|
||||
bool first;
|
||||
Buffer buf;
|
||||
List *w;
|
||||
MemoryContext tmpCtx;
|
||||
|
||||
@@ -259,7 +258,7 @@ typedef struct HnswVacuumState
|
||||
/* Methods */
|
||||
int HnswGetM(Relation index);
|
||||
int HnswGetEfConstruction(Relation index);
|
||||
FmgrInfo *HnswOptionalProcInfo(Relation rel, uint16 procnum);
|
||||
FmgrInfo *HnswOptionalProcInfo(Relation index, uint16 procnum);
|
||||
bool HnswNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
|
||||
void HnswCommitBuffer(Buffer buf, GenericXLogState *state);
|
||||
Buffer HnswNewBuffer(Relation index, ForkNumber forkNum);
|
||||
|
||||
@@ -81,7 +81,6 @@ HnswBuildAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **
|
||||
HnswPageGetOpaque(*page)->nextblkno = BufferGetBlockNumber(newbuf);
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(*buf);
|
||||
GenericXLogFinish(*state);
|
||||
UnlockReleaseBuffer(*buf);
|
||||
|
||||
@@ -179,7 +178,6 @@ CreateElementPages(HnswBuildState * buildstate)
|
||||
insertPage = BufferGetBlockNumber(buf);
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
UnlockReleaseBuffer(buf);
|
||||
|
||||
@@ -227,7 +225,6 @@ CreateNeighborPages(HnswBuildState * buildstate)
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
UnlockReleaseBuffer(buf);
|
||||
}
|
||||
|
||||
@@ -202,8 +202,6 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
|
||||
HnswInsertAppendPage(index, &newbuf, &newpage, state, page);
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(newbuf);
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
|
||||
/* Unlock previous buffer */
|
||||
@@ -270,9 +268,6 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
|
||||
}
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(buf);
|
||||
if (nbuf != buf)
|
||||
MarkBufferDirty(nbuf);
|
||||
GenericXLogFinish(state);
|
||||
UnlockReleaseBuffer(buf);
|
||||
if (nbuf != buf)
|
||||
@@ -391,7 +386,6 @@ HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswE
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
}
|
||||
else
|
||||
@@ -445,7 +439,6 @@ HnswAddDuplicate(Relation index, HnswElement element, HnswElement dup)
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
UnlockReleaseBuffer(buf);
|
||||
|
||||
|
||||
@@ -101,7 +101,6 @@ hnswbeginscan(Relation index, int nkeys, int norderbys)
|
||||
scan = RelationGetIndexScan(index, nkeys, norderbys);
|
||||
|
||||
so = (HnswScanOpaque) palloc(sizeof(HnswScanOpaqueData));
|
||||
so->buf = InvalidBuffer;
|
||||
so->first = true;
|
||||
so->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
|
||||
"Hnsw scan temporary context",
|
||||
@@ -182,7 +181,6 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
|
||||
{
|
||||
HnswCandidate *hc = llast(so->w);
|
||||
ItemPointer heaptid;
|
||||
BlockNumber indexblkno;
|
||||
|
||||
/* Move to next element if no valid heap TIDs */
|
||||
if (list_length(hc->element->heaptids) == 0)
|
||||
@@ -192,7 +190,6 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
|
||||
}
|
||||
|
||||
heaptid = llast(hc->element->heaptids);
|
||||
indexblkno = hc->element->blkno;
|
||||
|
||||
hc->element->heaptids = list_delete_last(hc->element->heaptids);
|
||||
|
||||
@@ -204,17 +201,14 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
|
||||
scan->xs_ctup.t_self = *heaptid;
|
||||
#endif
|
||||
|
||||
/* Unpin buffer */
|
||||
if (BufferIsValid(so->buf))
|
||||
ReleaseBuffer(so->buf);
|
||||
|
||||
/*
|
||||
* An index scan must maintain a pin on the index page holding the
|
||||
* item last returned by amgettuple
|
||||
* Typically, an index scan must maintain a pin on the index page
|
||||
* holding the item last returned by amgettuple. However, this is not
|
||||
* needed with the current vacuum strategy, which ensures scans do not
|
||||
* visit tuples in danger of being marked as deleted.
|
||||
*
|
||||
* https://www.postgresql.org/docs/current/index-locking.html
|
||||
*/
|
||||
so->buf = ReadBuffer(scan->indexRelation, indexblkno);
|
||||
|
||||
scan->xs_recheckorderby = false;
|
||||
return true;
|
||||
@@ -232,10 +226,6 @@ hnswendscan(IndexScanDesc scan)
|
||||
{
|
||||
HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
|
||||
|
||||
/* Release pin */
|
||||
if (BufferIsValid(so->buf))
|
||||
ReleaseBuffer(so->buf);
|
||||
|
||||
MemoryContextDelete(so->tmpCtx);
|
||||
|
||||
pfree(so);
|
||||
|
||||
@@ -38,12 +38,12 @@ HnswGetEfConstruction(Relation index)
|
||||
* Get proc
|
||||
*/
|
||||
FmgrInfo *
|
||||
HnswOptionalProcInfo(Relation rel, uint16 procnum)
|
||||
HnswOptionalProcInfo(Relation index, uint16 procnum)
|
||||
{
|
||||
if (!OidIsValid(index_getprocid(rel, 1, procnum)))
|
||||
if (!OidIsValid(index_getprocid(index, 1, procnum)))
|
||||
return NULL;
|
||||
|
||||
return index_getprocinfo(rel, 1, procnum);
|
||||
return index_getprocinfo(index, 1, procnum);
|
||||
}
|
||||
|
||||
/*
|
||||
@@ -117,7 +117,6 @@ HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState *
|
||||
void
|
||||
HnswCommitBuffer(Buffer buf, GenericXLogState *state)
|
||||
{
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
UnlockReleaseBuffer(buf);
|
||||
}
|
||||
@@ -143,6 +142,17 @@ HnswInitNeighbors(HnswElement element, int m)
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
* Free neighbors
|
||||
*/
|
||||
static void
|
||||
HnswFreeNeighbors(HnswElement element)
|
||||
{
|
||||
for (int lc = 0; lc <= element->level; lc++)
|
||||
pfree(element->neighbors[lc].items);
|
||||
pfree(element->neighbors);
|
||||
}
|
||||
|
||||
/*
|
||||
* Allocate an element
|
||||
*/
|
||||
@@ -174,10 +184,8 @@ HnswInitElement(ItemPointer heaptid, int m, double ml, int maxLevel)
|
||||
void
|
||||
HnswFreeElement(HnswElement element)
|
||||
{
|
||||
HnswFreeNeighbors(element);
|
||||
list_free_deep(element->heaptids);
|
||||
for (int lc = 0; lc <= element->level; lc++)
|
||||
pfree(element->neighbors[lc].items);
|
||||
pfree(element->neighbors);
|
||||
pfree(element->vec);
|
||||
pfree(element);
|
||||
}
|
||||
|
||||
@@ -128,10 +128,7 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
|
||||
blkno = HnswPageGetOpaque(page)->nextblkno;
|
||||
|
||||
if (updated)
|
||||
{
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
}
|
||||
else
|
||||
GenericXLogAbort(state);
|
||||
|
||||
@@ -229,7 +226,6 @@ RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswEleme
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
UnlockReleaseBuffer(buf);
|
||||
|
||||
@@ -330,7 +326,10 @@ RepairGraph(HnswVacuumState * vacuumstate)
|
||||
BufferAccessStrategy bas = vacuumstate->bas;
|
||||
BlockNumber blkno = HNSW_HEAD_BLKNO;
|
||||
|
||||
/* Wait for inserts to complete */
|
||||
/*
|
||||
* Wait for inserts to complete. Inserts before this point may have
|
||||
* neighbors about to be deleted. Inserts after this point will not.
|
||||
*/
|
||||
LockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
|
||||
UnlockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
|
||||
|
||||
@@ -443,7 +442,11 @@ MarkDeleted(HnswVacuumState * vacuumstate)
|
||||
Relation index = vacuumstate->index;
|
||||
BufferAccessStrategy bas = vacuumstate->bas;
|
||||
|
||||
/* Wait for selects to complete */
|
||||
/*
|
||||
* Wait for index scans to complete. Scans before this point may contain
|
||||
* tuples about to be deleted. Scans after this point will not, since the
|
||||
* graph has been repaired.
|
||||
*/
|
||||
LockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
|
||||
UnlockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
|
||||
|
||||
@@ -540,9 +543,6 @@ MarkDeleted(HnswVacuumState * vacuumstate)
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(buf);
|
||||
if (nbuf != buf)
|
||||
MarkBufferDirty(nbuf);
|
||||
GenericXLogFinish(state);
|
||||
if (nbuf != buf)
|
||||
UnlockReleaseBuffer(nbuf);
|
||||
|
||||
@@ -506,29 +506,30 @@ CreateListPages(Relation index, VectorArray centers, int dimensions,
|
||||
Buffer buf;
|
||||
Page page;
|
||||
GenericXLogState *state;
|
||||
OffsetNumber offno;
|
||||
Size itemsz;
|
||||
Size listSize;
|
||||
IvfflatList list;
|
||||
|
||||
itemsz = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
|
||||
list = palloc(itemsz);
|
||||
listSize = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
|
||||
list = palloc(listSize);
|
||||
|
||||
buf = IvfflatNewBuffer(index, forkNum);
|
||||
IvfflatInitRegisterPage(index, &buf, &page, &state);
|
||||
|
||||
for (int i = 0; i < lists; i++)
|
||||
{
|
||||
OffsetNumber offno;
|
||||
|
||||
/* Load list */
|
||||
list->startPage = InvalidBlockNumber;
|
||||
list->insertPage = InvalidBlockNumber;
|
||||
memcpy(&list->center, VectorArrayGet(centers, i), VECTOR_SIZE(dimensions));
|
||||
|
||||
/* Ensure free space */
|
||||
if (PageGetFreeSpace(page) < itemsz)
|
||||
if (PageGetFreeSpace(page) < listSize)
|
||||
IvfflatAppendPage(index, &buf, &page, &state, forkNum);
|
||||
|
||||
/* Add the item */
|
||||
offno = PageAddItem(page, (Item) list, itemsz, InvalidOffsetNumber, false, false);
|
||||
offno = PageAddItem(page, (Item) list, listSize, InvalidOffsetNumber, false, false);
|
||||
if (offno == InvalidOffsetNumber)
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
|
||||
@@ -182,15 +182,6 @@ ivfflatvalidate(Oid opclassoid)
|
||||
return true;
|
||||
}
|
||||
|
||||
/*
|
||||
* Checks if index-only scan is supported
|
||||
*/
|
||||
static bool
|
||||
ivfflatcanreturn(Relation index, int attno)
|
||||
{
|
||||
return attno == 1 && IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC) == NULL;
|
||||
}
|
||||
|
||||
/*
|
||||
* Define index handler
|
||||
*
|
||||
@@ -232,7 +223,7 @@ ivfflathandler(PG_FUNCTION_ARGS)
|
||||
amroutine->aminsert = ivfflatinsert;
|
||||
amroutine->ambulkdelete = ivfflatbulkdelete;
|
||||
amroutine->amvacuumcleanup = ivfflatvacuumcleanup;
|
||||
amroutine->amcanreturn = ivfflatcanreturn;
|
||||
amroutine->amcanreturn = NULL; /* tuple not included in heapsort */
|
||||
amroutine->amcostestimate = ivfflatcostestimate;
|
||||
amroutine->amoptions = ivfflatoptions;
|
||||
amroutine->amproperty = NULL; /* TODO AMPROP_DISTANCE_ORDERABLE */
|
||||
|
||||
@@ -248,7 +248,6 @@ typedef struct IvfflatScanOpaqueData
|
||||
bool first;
|
||||
Buffer buf;
|
||||
ItemPointerData heaptid;
|
||||
IndexTuple itup;
|
||||
|
||||
/* Sorting */
|
||||
Tuplesortstate *sortstate;
|
||||
@@ -278,7 +277,7 @@ VectorArray VectorArrayInit(int maxlen, int dimensions);
|
||||
void VectorArrayFree(VectorArray arr);
|
||||
void PrintVectorArray(char *msg, VectorArray arr);
|
||||
void IvfflatKmeans(Relation index, VectorArray samples, VectorArray centers);
|
||||
FmgrInfo *IvfflatOptionalProcInfo(Relation rel, uint16 procnum);
|
||||
FmgrInfo *IvfflatOptionalProcInfo(Relation index, uint16 procnum);
|
||||
bool IvfflatNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
|
||||
int IvfflatGetLists(Relation index);
|
||||
void IvfflatGetMetaPageInfo(Relation index, int *lists, int *dimensions);
|
||||
|
||||
@@ -11,36 +11,37 @@
|
||||
* Find the list that minimizes the distance function
|
||||
*/
|
||||
static void
|
||||
FindInsertPage(Relation rel, Datum *values, BlockNumber *insertPage, ListInfo * listInfo)
|
||||
FindInsertPage(Relation index, Datum *values, BlockNumber *insertPage, ListInfo * listInfo)
|
||||
{
|
||||
Buffer cbuf;
|
||||
Page cpage;
|
||||
IvfflatList list;
|
||||
double distance;
|
||||
double minDistance = DBL_MAX;
|
||||
BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
|
||||
FmgrInfo *procinfo;
|
||||
Oid collation;
|
||||
OffsetNumber offno;
|
||||
OffsetNumber maxoffno;
|
||||
|
||||
/* Avoid compiler warning */
|
||||
listInfo->blkno = nextblkno;
|
||||
listInfo->offno = FirstOffsetNumber;
|
||||
|
||||
procinfo = index_getprocinfo(rel, 1, IVFFLAT_DISTANCE_PROC);
|
||||
collation = rel->rd_indcollation[0];
|
||||
procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC);
|
||||
collation = index->rd_indcollation[0];
|
||||
|
||||
/* Search all list pages */
|
||||
while (BlockNumberIsValid(nextblkno))
|
||||
{
|
||||
cbuf = ReadBuffer(rel, nextblkno);
|
||||
Buffer cbuf;
|
||||
Page cpage;
|
||||
OffsetNumber maxoffno;
|
||||
|
||||
cbuf = ReadBuffer(index, nextblkno);
|
||||
LockBuffer(cbuf, BUFFER_LOCK_SHARE);
|
||||
cpage = BufferGetPage(cbuf);
|
||||
maxoffno = PageGetMaxOffsetNumber(cpage);
|
||||
|
||||
for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
|
||||
for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
|
||||
{
|
||||
IvfflatList list;
|
||||
double distance;
|
||||
|
||||
list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
|
||||
distance = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, values[0], PointerGetDatum(&list->center)));
|
||||
|
||||
@@ -63,7 +64,7 @@ FindInsertPage(Relation rel, Datum *values, BlockNumber *insertPage, ListInfo *
|
||||
* Insert a tuple into the index
|
||||
*/
|
||||
static void
|
||||
InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
|
||||
InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
|
||||
{
|
||||
IndexTuple itup;
|
||||
Datum value;
|
||||
@@ -80,20 +81,20 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
|
||||
value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
|
||||
|
||||
/* Normalize if needed */
|
||||
normprocinfo = IvfflatOptionalProcInfo(rel, IVFFLAT_NORM_PROC);
|
||||
normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC);
|
||||
if (normprocinfo != NULL)
|
||||
{
|
||||
if (!IvfflatNormValue(normprocinfo, rel->rd_indcollation[0], &value, NULL))
|
||||
if (!IvfflatNormValue(normprocinfo, index->rd_indcollation[0], &value, NULL))
|
||||
return;
|
||||
}
|
||||
|
||||
/* Find the insert page - sets the page and list info */
|
||||
FindInsertPage(rel, values, &insertPage, &listInfo);
|
||||
FindInsertPage(index, values, &insertPage, &listInfo);
|
||||
Assert(BlockNumberIsValid(insertPage));
|
||||
originalInsertPage = insertPage;
|
||||
|
||||
/* Form tuple */
|
||||
itup = index_form_tuple(RelationGetDescr(rel), &value, isnull);
|
||||
itup = index_form_tuple(RelationGetDescr(index), &value, isnull);
|
||||
itup->t_tid = *heap_tid;
|
||||
|
||||
/* Get tuple size */
|
||||
@@ -103,10 +104,10 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
|
||||
/* Find a page to insert the item */
|
||||
for (;;)
|
||||
{
|
||||
buf = ReadBuffer(rel, insertPage);
|
||||
buf = ReadBuffer(index, insertPage);
|
||||
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
|
||||
|
||||
state = GenericXLogStart(rel);
|
||||
state = GenericXLogStart(index);
|
||||
page = GenericXLogRegisterBuffer(state, buf, 0);
|
||||
|
||||
if (PageGetFreeSpace(page) >= itemsz)
|
||||
@@ -126,9 +127,9 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
|
||||
Page newpage;
|
||||
|
||||
/* Add a new page */
|
||||
LockRelationForExtension(rel, ExclusiveLock);
|
||||
newbuf = IvfflatNewBuffer(rel, MAIN_FORKNUM);
|
||||
UnlockRelationForExtension(rel, ExclusiveLock);
|
||||
LockRelationForExtension(index, ExclusiveLock);
|
||||
newbuf = IvfflatNewBuffer(index, MAIN_FORKNUM);
|
||||
UnlockRelationForExtension(index, ExclusiveLock);
|
||||
|
||||
/* Init new page */
|
||||
newpage = GenericXLogRegisterBuffer(state, newbuf, GENERIC_XLOG_FULL_IMAGE);
|
||||
@@ -141,15 +142,13 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
|
||||
IvfflatPageGetOpaque(page)->nextblkno = insertPage;
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(newbuf);
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
|
||||
/* Unlock previous buffer */
|
||||
UnlockReleaseBuffer(buf);
|
||||
|
||||
/* Prepare new buffer */
|
||||
state = GenericXLogStart(rel);
|
||||
state = GenericXLogStart(index);
|
||||
buf = newbuf;
|
||||
page = GenericXLogRegisterBuffer(state, buf, 0);
|
||||
break;
|
||||
@@ -158,13 +157,13 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
|
||||
|
||||
/* Add to next offset */
|
||||
if (PageAddItem(page, (Item) itup, itemsz, InvalidOffsetNumber, false, false) == InvalidOffsetNumber)
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(rel));
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
IvfflatCommitBuffer(buf, state);
|
||||
|
||||
/* Update the insert page */
|
||||
if (insertPage != originalInsertPage)
|
||||
IvfflatUpdateList(rel, listInfo, insertPage, originalInsertPage, InvalidBlockNumber, MAIN_FORKNUM);
|
||||
IvfflatUpdateList(index, listInfo, insertPage, originalInsertPage, InvalidBlockNumber, MAIN_FORKNUM);
|
||||
}
|
||||
|
||||
/*
|
||||
|
||||
@@ -17,10 +17,6 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
|
||||
FmgrInfo *procinfo;
|
||||
Oid collation;
|
||||
int64 j;
|
||||
double distance;
|
||||
double sum;
|
||||
double choice;
|
||||
Vector *vec;
|
||||
float *weight = palloc(samples->length * sizeof(float));
|
||||
int numCenters = centers->maxlen;
|
||||
int numSamples = samples->length;
|
||||
@@ -33,17 +29,21 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
|
||||
centers->length++;
|
||||
|
||||
for (j = 0; j < numSamples; j++)
|
||||
weight[j] = DBL_MAX;
|
||||
weight[j] = FLT_MAX;
|
||||
|
||||
for (int i = 0; i < numCenters; i++)
|
||||
{
|
||||
double sum;
|
||||
double choice;
|
||||
|
||||
CHECK_FOR_INTERRUPTS();
|
||||
|
||||
sum = 0.0;
|
||||
|
||||
for (j = 0; j < numSamples; j++)
|
||||
{
|
||||
vec = VectorArrayGet(samples, j);
|
||||
Vector *vec = VectorArrayGet(samples, j);
|
||||
double distance;
|
||||
|
||||
/* Only need to compute distance for new center */
|
||||
/* TODO Use triangle inequality to reduce distance calculations */
|
||||
@@ -112,7 +112,6 @@ CompareVectors(const void *a, const void *b)
|
||||
static void
|
||||
QuickCenters(Relation index, VectorArray samples, VectorArray centers)
|
||||
{
|
||||
Vector *vec;
|
||||
int dimensions = centers->dim;
|
||||
Oid collation = index->rd_indcollation[0];
|
||||
FmgrInfo *normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_KMEANS_NORM_PROC);
|
||||
@@ -123,7 +122,7 @@ QuickCenters(Relation index, VectorArray samples, VectorArray centers)
|
||||
qsort(samples->items, samples->length, VECTOR_SIZE(samples->dim), CompareVectors);
|
||||
for (int i = 0; i < samples->length; i++)
|
||||
{
|
||||
vec = VectorArrayGet(samples, i);
|
||||
Vector *vec = VectorArrayGet(samples, i);
|
||||
|
||||
if (i == 0 || CompareVectors(vec, VectorArrayGet(samples, i - 1)) != 0)
|
||||
{
|
||||
@@ -136,7 +135,7 @@ QuickCenters(Relation index, VectorArray samples, VectorArray centers)
|
||||
/* Fill remaining with random data */
|
||||
while (centers->length < centers->maxlen)
|
||||
{
|
||||
vec = VectorArrayGet(centers, centers->length);
|
||||
Vector *vec = VectorArrayGet(centers, centers->length);
|
||||
|
||||
SET_VARSIZE(vec, VECTOR_SIZE(dimensions));
|
||||
vec->dim = dimensions;
|
||||
@@ -168,7 +167,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
Oid collation;
|
||||
Vector *vec;
|
||||
Vector *newCenter;
|
||||
int iteration;
|
||||
int64 j;
|
||||
int64 k;
|
||||
int dimensions = centers->dim;
|
||||
@@ -182,14 +180,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
float *s;
|
||||
float *halfcdist;
|
||||
float *newcdist;
|
||||
int changes;
|
||||
double minDistance;
|
||||
int closestCenter;
|
||||
double distance;
|
||||
bool rj;
|
||||
bool rjreset;
|
||||
double dxcx;
|
||||
double dxc;
|
||||
|
||||
/* Calculate allocation sizes */
|
||||
Size samplesSize = VECTOR_ARRAY_SIZE(samples->maxlen, samples->dim);
|
||||
@@ -247,14 +237,14 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
/* Assign each x to its closest initial center c(x) = argmin d(x,c) */
|
||||
for (j = 0; j < numSamples; j++)
|
||||
{
|
||||
minDistance = DBL_MAX;
|
||||
closestCenter = 0;
|
||||
float minDistance = FLT_MAX;
|
||||
int closestCenter = 0;
|
||||
|
||||
/* Find closest center */
|
||||
for (k = 0; k < numCenters; k++)
|
||||
{
|
||||
/* TODO Use Lemma 1 in k-means++ initialization */
|
||||
distance = lowerBound[j * numCenters + k];
|
||||
float distance = lowerBound[j * numCenters + k];
|
||||
|
||||
if (distance < minDistance)
|
||||
{
|
||||
@@ -268,13 +258,14 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
}
|
||||
|
||||
/* Give 500 iterations to converge */
|
||||
for (iteration = 0; iteration < 500; iteration++)
|
||||
for (int iteration = 0; iteration < 500; iteration++)
|
||||
{
|
||||
int changes = 0;
|
||||
bool rjreset;
|
||||
|
||||
/* Can take a while, so ensure we can interrupt */
|
||||
CHECK_FOR_INTERRUPTS();
|
||||
|
||||
changes = 0;
|
||||
|
||||
/* Step 1: For all centers, compute distance */
|
||||
for (j = 0; j < numCenters; j++)
|
||||
{
|
||||
@@ -282,7 +273,8 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
|
||||
for (k = j + 1; k < numCenters; k++)
|
||||
{
|
||||
distance = 0.5 * DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
|
||||
float distance = 0.5 * DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
|
||||
|
||||
halfcdist[j * numCenters + k] = distance;
|
||||
halfcdist[k * numCenters + j] = distance;
|
||||
}
|
||||
@@ -291,10 +283,12 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
/* For all centers c, compute s(c) */
|
||||
for (j = 0; j < numCenters; j++)
|
||||
{
|
||||
minDistance = DBL_MAX;
|
||||
float minDistance = FLT_MAX;
|
||||
|
||||
for (k = 0; k < numCenters; k++)
|
||||
{
|
||||
float distance;
|
||||
|
||||
if (j == k)
|
||||
continue;
|
||||
|
||||
@@ -310,6 +304,8 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
|
||||
for (j = 0; j < numSamples; j++)
|
||||
{
|
||||
bool rj;
|
||||
|
||||
/* Step 2: Identify all points x such that u(x) <= s(c(x)) */
|
||||
if (upperBound[j] <= s[closestCenters[j]])
|
||||
continue;
|
||||
@@ -318,6 +314,8 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
|
||||
for (k = 0; k < numCenters; k++)
|
||||
{
|
||||
float dxcx;
|
||||
|
||||
/* Step 3: For all remaining points x and centers c */
|
||||
if (k == closestCenters[j])
|
||||
continue;
|
||||
@@ -347,7 +345,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
/* Step 3b */
|
||||
if (dxcx > lowerBound[j * numCenters + k] || dxcx > halfcdist[closestCenters[j] * numCenters + k])
|
||||
{
|
||||
dxc = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
|
||||
float dxc = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
|
||||
|
||||
/* d(x,c) calculated */
|
||||
lowerBound[j * numCenters + k] = dxc;
|
||||
@@ -361,7 +359,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
|
||||
changes++;
|
||||
}
|
||||
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -378,6 +375,8 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
|
||||
for (j = 0; j < numSamples; j++)
|
||||
{
|
||||
int closestCenter;
|
||||
|
||||
vec = VectorArrayGet(samples, j);
|
||||
closestCenter = closestCenters[j];
|
||||
|
||||
@@ -426,7 +425,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
{
|
||||
for (k = 0; k < numCenters; k++)
|
||||
{
|
||||
distance = lowerBound[j * numCenters + k] - newcdist[k];
|
||||
float distance = lowerBound[j * numCenters + k] - newcdist[k];
|
||||
|
||||
if (distance < 0)
|
||||
distance = 0;
|
||||
@@ -442,7 +441,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
|
||||
|
||||
/* Step 7 */
|
||||
for (j = 0; j < numCenters; j++)
|
||||
memcpy(VectorArrayGet(centers, j), VectorArrayGet(newCenters, j), VECTOR_SIZE(dimensions));
|
||||
VectorArraySet(centers, j, VectorArrayGet(newCenters, j));
|
||||
|
||||
if (changes == 0 && iteration != 0)
|
||||
break;
|
||||
@@ -465,9 +464,6 @@ static void
|
||||
CheckCenters(Relation index, VectorArray centers)
|
||||
{
|
||||
FmgrInfo *normprocinfo;
|
||||
Oid collation;
|
||||
Vector *vec;
|
||||
double norm;
|
||||
|
||||
if (centers->length != centers->maxlen)
|
||||
elog(ERROR, "Not enough centers. Please report a bug.");
|
||||
@@ -475,7 +471,7 @@ CheckCenters(Relation index, VectorArray centers)
|
||||
/* Ensure no NaN or infinite values */
|
||||
for (int i = 0; i < centers->length; i++)
|
||||
{
|
||||
vec = VectorArrayGet(centers, i);
|
||||
Vector *vec = VectorArrayGet(centers, i);
|
||||
|
||||
for (int j = 0; j < vec->dim; j++)
|
||||
{
|
||||
@@ -501,11 +497,12 @@ CheckCenters(Relation index, VectorArray centers)
|
||||
normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC);
|
||||
if (normprocinfo != NULL)
|
||||
{
|
||||
collation = index->rd_indcollation[0];
|
||||
Oid collation = index->rd_indcollation[0];
|
||||
|
||||
for (int i = 0; i < centers->length; i++)
|
||||
{
|
||||
norm = DatumGetFloat8(FunctionCall1Coll(normprocinfo, collation, PointerGetDatum(VectorArrayGet(centers, i))));
|
||||
double norm = DatumGetFloat8(FunctionCall1Coll(normprocinfo, collation, PointerGetDatum(VectorArrayGet(centers, i))));
|
||||
|
||||
if (norm == 0)
|
||||
elog(ERROR, "Zero norm detected. Please report a bug.");
|
||||
}
|
||||
|
||||
@@ -141,7 +141,6 @@ GetScanItems(IndexScanDesc scan, Datum value)
|
||||
IndexTuple itup;
|
||||
Datum datum;
|
||||
bool isnull;
|
||||
ItemPointerData indextid;
|
||||
ItemId itemid = PageGetItemId(page, offno);
|
||||
|
||||
/* Skip dead tuples */
|
||||
@@ -150,7 +149,6 @@ GetScanItems(IndexScanDesc scan, Datum value)
|
||||
|
||||
itup = (IndexTuple) PageGetItem(page, itemid);
|
||||
datum = index_getattr(itup, 1, tupdesc, &isnull);
|
||||
ItemPointerSet(&indextid, searchPage, offno);
|
||||
|
||||
/*
|
||||
* Add virtual tuple
|
||||
@@ -163,7 +161,7 @@ GetScanItems(IndexScanDesc scan, Datum value)
|
||||
slot->tts_isnull[0] = false;
|
||||
slot->tts_values[1] = PointerGetDatum(&itup->t_tid);
|
||||
slot->tts_isnull[1] = false;
|
||||
slot->tts_values[2] = PointerGetDatum(&indextid);
|
||||
slot->tts_values[2] = Int32GetDatum((int) searchPage);
|
||||
slot->tts_isnull[2] = false;
|
||||
ExecStoreVirtualTuple(slot);
|
||||
|
||||
@@ -238,34 +236,6 @@ MarkPriorTupleDead(IndexScanDesc scan)
|
||||
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
|
||||
}
|
||||
|
||||
/*
|
||||
* Set tuple for index-only scan
|
||||
*/
|
||||
static void
|
||||
SetIndexTuple(IndexScanDesc scan, ItemPointer indextid)
|
||||
{
|
||||
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
|
||||
Buffer buf = so->buf;
|
||||
Page page;
|
||||
OffsetNumber offno = ItemPointerGetOffsetNumber(indextid);
|
||||
IndexTuple itup;
|
||||
Size itupSize;
|
||||
|
||||
LockBuffer(buf, BUFFER_LOCK_SHARE);
|
||||
page = BufferGetPage(buf);
|
||||
itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offno));
|
||||
itupSize = IndexTupleSize(itup);
|
||||
|
||||
if (so->itup == NULL)
|
||||
so->itup = palloc(BLCKSZ);
|
||||
|
||||
memcpy(so->itup, itup, itupSize);
|
||||
|
||||
scan->xs_itup = so->itup;
|
||||
|
||||
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
|
||||
}
|
||||
|
||||
/*
|
||||
* Prepare for an index scan
|
||||
*/
|
||||
@@ -294,7 +264,6 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
|
||||
so->buf = InvalidBuffer;
|
||||
so->first = true;
|
||||
ItemPointerSetInvalid(&so->heaptid);
|
||||
so->itup = NULL;
|
||||
so->probes = probes;
|
||||
so->dimensions = dimensions;
|
||||
|
||||
@@ -311,7 +280,7 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
|
||||
#endif
|
||||
TupleDescInitEntry(so->tupdesc, (AttrNumber) 1, "distance", FLOAT8OID, -1, 0);
|
||||
TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "heaptid", TIDOID, -1, 0);
|
||||
TupleDescInitEntry(so->tupdesc, (AttrNumber) 3, "indextid", TIDOID, -1, 0);
|
||||
TupleDescInitEntry(so->tupdesc, (AttrNumber) 3, "indexblkno", INT4OID, -1, 0);
|
||||
|
||||
/* Prep sort */
|
||||
so->sortstate = tuplesort_begin_heap(so->tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, work_mem, NULL, false);
|
||||
@@ -326,8 +295,6 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
|
||||
|
||||
scan->opaque = so;
|
||||
|
||||
scan->xs_itupdesc = RelationGetDescr(index);
|
||||
|
||||
return scan;
|
||||
}
|
||||
|
||||
@@ -413,7 +380,7 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
|
||||
if (tuplesort_gettupleslot(so->sortstate, true, false, so->slot, NULL))
|
||||
{
|
||||
ItemPointer heaptid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
|
||||
ItemPointer indextid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 3, &so->isnull));
|
||||
BlockNumber indexblkno = DatumGetInt32(slot_getattr(so->slot, 3, &so->isnull));
|
||||
|
||||
#if PG_VERSION_NUM >= 120000
|
||||
scan->xs_heaptid = *heaptid;
|
||||
@@ -434,11 +401,7 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
|
||||
*
|
||||
* https://www.postgresql.org/docs/current/index-locking.html
|
||||
*/
|
||||
so->buf = ReadBuffer(scan->indexRelation, ItemPointerGetBlockNumber(indextid));
|
||||
|
||||
/* Set tuple for index-only scan */
|
||||
if (scan->xs_want_itup)
|
||||
SetIndexTuple(scan, indextid);
|
||||
so->buf = ReadBuffer(scan->indexRelation, indexblkno);
|
||||
|
||||
scan->xs_recheckorderby = false;
|
||||
return true;
|
||||
@@ -459,9 +422,6 @@ ivfflatendscan(IndexScanDesc scan)
|
||||
if (BufferIsValid(so->buf))
|
||||
ReleaseBuffer(so->buf);
|
||||
|
||||
if (so->itup != NULL)
|
||||
pfree(so->itup);
|
||||
|
||||
pairingheap_free(so->listQueue);
|
||||
tuplesort_end(so->sortstate);
|
||||
|
||||
|
||||
@@ -57,12 +57,12 @@ IvfflatGetLists(Relation index)
|
||||
* Get proc
|
||||
*/
|
||||
FmgrInfo *
|
||||
IvfflatOptionalProcInfo(Relation rel, uint16 procnum)
|
||||
IvfflatOptionalProcInfo(Relation index, uint16 procnum)
|
||||
{
|
||||
if (!OidIsValid(index_getprocid(rel, 1, procnum)))
|
||||
if (!OidIsValid(index_getprocid(index, 1, procnum)))
|
||||
return NULL;
|
||||
|
||||
return index_getprocinfo(rel, 1, procnum);
|
||||
return index_getprocinfo(index, 1, procnum);
|
||||
}
|
||||
|
||||
/*
|
||||
@@ -136,7 +136,6 @@ IvfflatInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogStat
|
||||
void
|
||||
IvfflatCommitBuffer(Buffer buf, GenericXLogState *state)
|
||||
{
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
UnlockReleaseBuffer(buf);
|
||||
}
|
||||
@@ -160,8 +159,6 @@ IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **st
|
||||
IvfflatInitPage(newbuf, newpage);
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(*buf);
|
||||
MarkBufferDirty(newbuf);
|
||||
GenericXLogFinish(*state);
|
||||
|
||||
/* Unlock */
|
||||
|
||||
@@ -107,7 +107,6 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
|
||||
{
|
||||
/* Delete tuples */
|
||||
PageIndexMultiDelete(page, deletable, ndeletable);
|
||||
MarkBufferDirty(buf);
|
||||
GenericXLogFinish(state);
|
||||
}
|
||||
else
|
||||
|
||||
Reference in New Issue
Block a user