static void
spgRedoVacuumLeaf(XLogRecPtr lsn, XLogRecord *record)
{
char *ptr = XLogRecGetData(record);
spgxlogVacuumLeaf *xldata = (spgxlogVacuumLeaf *) ptr;
OffsetNumber *toDead;
OffsetNumber *toPlaceholder;
OffsetNumber *moveSrc;
OffsetNumber *moveDest;
OffsetNumber *chainSrc;
OffsetNumber *chainDest;
SpGistState state;
Buffer buffer;
Page page;
int i;
fillFakeState(&state, xldata->stateSrc);
ptr += sizeof(spgxlogVacuumLeaf);
toDead = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nDead;
toPlaceholder = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nPlaceholder;
moveSrc = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nMove;
moveDest = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nMove;
chainSrc = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nChain;
chainDest = (OffsetNumber *) ptr;
if (!(record->xl_info & XLR_BKP_BLOCK_1))
{
buffer = XLogReadBuffer(xldata->node, xldata->blkno, false);
if (BufferIsValid(buffer))
{
page = BufferGetPage(buffer);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
spgPageIndexMultiDelete(&state, page,
toDead, xldata->nDead,
SPGIST_DEAD, SPGIST_DEAD,
InvalidBlockNumber,
InvalidOffsetNumber);
spgPageIndexMultiDelete(&state, page,
toPlaceholder, xldata->nPlaceholder,
SPGIST_PLACEHOLDER, SPGIST_PLACEHOLDER,
InvalidBlockNumber,
InvalidOffsetNumber);
/* see comments in vacuumLeafPage() */
for (i = 0; i < xldata->nMove; i++)
{
ItemId idSrc = PageGetItemId(page, moveSrc[i]);
ItemId idDest = PageGetItemId(page, moveDest[i]);
ItemIdData tmp;
tmp = *idSrc;
*idSrc = *idDest;
*idDest = tmp;
}
spgPageIndexMultiDelete(&state, page,
moveSrc, xldata->nMove,
SPGIST_PLACEHOLDER, SPGIST_PLACEHOLDER,
InvalidBlockNumber,
InvalidOffsetNumber);
for (i = 0; i < xldata->nChain; i++)
{
SpGistLeafTuple lt;
lt = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, chainSrc[i]));
Assert(lt->tupstate == SPGIST_LIVE);
lt->nextOffset = chainDest[i];
}
PageSetLSN(page, lsn);
PageSetTLI(page, ThisTimeLineID);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
}
}
/*
* Vacuum a regular (non-root) leaf page
*
* We must delete tuples that are targeted for deletion by the VACUUM,
* but not move any tuples that are referenced by outside links; we assume
* those are the ones that are heads of chains.
*
* If we find a REDIRECT that was made by a concurrently-running transaction,
* we must add its target TID to pendingList. (We don't try to visit the
* target immediately, first because we don't want VACUUM locking more than
* one buffer at a time, and second because the duplicate-filtering logic
* in spgAddPendingTID is useful to ensure we can't get caught in an infinite
* loop in the face of continuous concurrent insertions.)
*
* If forPending is true, we are examining the page as a consequence of
* chasing a redirect link, not as part of the normal sequential scan.
* We still vacuum the page normally, but we don't increment the stats
* about live tuples; else we'd double-count those tuples, since the page
* has been or will be visited in the sequential scan as well.
*/
static void
vacuumLeafPage(spgBulkDeleteState *bds, Relation index, Buffer buffer,
bool forPending)
{
Page page = BufferGetPage(buffer);
spgxlogVacuumLeaf xlrec;
XLogRecData rdata[8];
OffsetNumber toDead[MaxIndexTuplesPerPage];
OffsetNumber toPlaceholder[MaxIndexTuplesPerPage];
OffsetNumber moveSrc[MaxIndexTuplesPerPage];
OffsetNumber moveDest[MaxIndexTuplesPerPage];
OffsetNumber chainSrc[MaxIndexTuplesPerPage];
OffsetNumber chainDest[MaxIndexTuplesPerPage];
OffsetNumber predecessor[MaxIndexTuplesPerPage + 1];
bool deletable[MaxIndexTuplesPerPage + 1];
int nDeletable;
OffsetNumber i,
max = PageGetMaxOffsetNumber(page);
memset(predecessor, 0, sizeof(predecessor));
memset(deletable, 0, sizeof(deletable));
nDeletable = 0;
/* Scan page, identify tuples to delete, accumulate stats */
for (i = FirstOffsetNumber; i <= max; i++)
{
SpGistLeafTuple lt;
lt = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, i));
if (lt->tupstate == SPGIST_LIVE)
{
Assert(ItemPointerIsValid(<->heapPtr));
if (bds->callback(<->heapPtr, bds->callback_state))
{
bds->stats->tuples_removed += 1;
deletable[i] = true;
nDeletable++;
}
else
{
if (!forPending)
bds->stats->num_index_tuples += 1;
}
/* Form predecessor map, too */
if (lt->nextOffset != InvalidOffsetNumber)
{
/* paranoia about corrupted chain links */
if (lt->nextOffset < FirstOffsetNumber ||
lt->nextOffset > max ||
predecessor[lt->nextOffset] != InvalidOffsetNumber)
elog(ERROR, "inconsistent tuple chain links in page %u of index \"%s\"",
BufferGetBlockNumber(buffer),
RelationGetRelationName(index));
predecessor[lt->nextOffset] = i;
}
}
else if (lt->tupstate == SPGIST_REDIRECT)
{
SpGistDeadTuple dt = (SpGistDeadTuple) lt;
Assert(dt->nextOffset == InvalidOffsetNumber);
Assert(ItemPointerIsValid(&dt->pointer));
/*
* Add target TID to pending list if the redirection could have
* happened since VACUUM started.
*
* Note: we could make a tighter test by seeing if the xid is
* "running" according to the active snapshot; but tqual.c doesn't
* currently export a suitable API, and it's not entirely clear
* that a tighter test is worth the cycles anyway.
*/
if (TransactionIdFollowsOrEquals(dt->xid, bds->myXmin))
spgAddPendingTID(bds, &dt->pointer);
}
else
{
Assert(lt->nextOffset == InvalidOffsetNumber);
}
}
if (nDeletable == 0)
return; /* nothing more to do */
/*----------
* Figure out exactly what we have to do. We do this separately from
* actually modifying the page, mainly so that we have a representation
* that can be dumped into WAL and then the replay code can do exactly
* the same thing. The output of this step consists of six arrays
* describing four kinds of operations, to be performed in this order:
*
* toDead[]: tuple numbers to be replaced with DEAD tuples
* toPlaceholder[]: tuple numbers to be replaced with PLACEHOLDER tuples
* moveSrc[]: tuple numbers that need to be relocated to another offset
* (replacing the tuple there) and then replaced with PLACEHOLDER tuples
* moveDest[]: new locations for moveSrc tuples
* chainSrc[]: tuple numbers whose chain links (nextOffset) need updates
* chainDest[]: new values of nextOffset for chainSrc members
*
* It's easiest to figure out what we have to do by processing tuple
* chains, so we iterate over all the tuples (not just the deletable
* ones!) to identify chain heads, then chase down each chain and make
* work item entries for deletable tuples within the chain.
*----------
*/
xlrec.nDead = xlrec.nPlaceholder = xlrec.nMove = xlrec.nChain = 0;
for (i = FirstOffsetNumber; i <= max; i++)
{
SpGistLeafTuple head;
bool interveningDeletable;
OffsetNumber prevLive;
OffsetNumber j;
head = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, i));
if (head->tupstate != SPGIST_LIVE)
continue; /* can't be a chain member */
if (predecessor[i] != 0)
continue; /* not a chain head */
/* initialize ... */
interveningDeletable = false;
prevLive = deletable[i] ? InvalidOffsetNumber : i;
/* scan down the chain ... */
j = head->nextOffset;
while (j != InvalidOffsetNumber)
{
SpGistLeafTuple lt;
lt = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, j));
if (lt->tupstate != SPGIST_LIVE)
{
/* all tuples in chain should be live */
elog(ERROR, "unexpected SPGiST tuple state: %d",
lt->tupstate);
}
if (deletable[j])
{
/* This tuple should be replaced by a placeholder */
toPlaceholder[xlrec.nPlaceholder] = j;
xlrec.nPlaceholder++;
/* previous live tuple's chain link will need an update */
interveningDeletable = true;
}
else if (prevLive == InvalidOffsetNumber)
{
/*
* This is the first live tuple in the chain. It has to move
* to the head position.
*/
moveSrc[xlrec.nMove] = j;
moveDest[xlrec.nMove] = i;
xlrec.nMove++;
/* Chain updates will be applied after the move */
prevLive = i;
interveningDeletable = false;
}
else
{
/*
* Second or later live tuple. Arrange to re-chain it to the
* previous live one, if there was a gap.
*/
if (interveningDeletable)
{
chainSrc[xlrec.nChain] = prevLive;
chainDest[xlrec.nChain] = j;
xlrec.nChain++;
}
prevLive = j;
interveningDeletable = false;
}
j = lt->nextOffset;
}
if (prevLive == InvalidOffsetNumber)
{
/* The chain is entirely removable, so we need a DEAD tuple */
toDead[xlrec.nDead] = i;
xlrec.nDead++;
}
else if (interveningDeletable)
{
/* One or more deletions at end of chain, so close it off */
chainSrc[xlrec.nChain] = prevLive;
chainDest[xlrec.nChain] = InvalidOffsetNumber;
xlrec.nChain++;
}
}
/* sanity check ... */
if (nDeletable != xlrec.nDead + xlrec.nPlaceholder + xlrec.nMove)
elog(ERROR, "inconsistent counts of deletable tuples");
/* Prepare WAL record */
xlrec.node = index->rd_node;
xlrec.blkno = BufferGetBlockNumber(buffer);
STORE_STATE(&bds->spgstate, xlrec.stateSrc);
ACCEPT_RDATA_DATA(&xlrec, SizeOfSpgxlogVacuumLeaf, 0);
ACCEPT_RDATA_DATA(toDead, sizeof(OffsetNumber) * xlrec.nDead, 1);
ACCEPT_RDATA_DATA(toPlaceholder, sizeof(OffsetNumber) * xlrec.nPlaceholder, 2);
ACCEPT_RDATA_DATA(moveSrc, sizeof(OffsetNumber) * xlrec.nMove, 3);
ACCEPT_RDATA_DATA(moveDest, sizeof(OffsetNumber) * xlrec.nMove, 4);
ACCEPT_RDATA_DATA(chainSrc, sizeof(OffsetNumber) * xlrec.nChain, 5);
ACCEPT_RDATA_DATA(chainDest, sizeof(OffsetNumber) * xlrec.nChain, 6);
ACCEPT_RDATA_BUFFER(buffer, 7);
/* Do the updates */
START_CRIT_SECTION();
spgPageIndexMultiDelete(&bds->spgstate, page,
toDead, xlrec.nDead,
SPGIST_DEAD, SPGIST_DEAD,
InvalidBlockNumber, InvalidOffsetNumber);
spgPageIndexMultiDelete(&bds->spgstate, page,
toPlaceholder, xlrec.nPlaceholder,
SPGIST_PLACEHOLDER, SPGIST_PLACEHOLDER,
InvalidBlockNumber, InvalidOffsetNumber);
/*
* We implement the move step by swapping the item pointers of the source
* and target tuples, then replacing the newly-source tuples with
* placeholders. This is perhaps unduly friendly with the page data
* representation, but it's fast and doesn't risk page overflow when a
* tuple to be relocated is large.
*/
for (i = 0; i < xlrec.nMove; i++)
{
ItemId idSrc = PageGetItemId(page, moveSrc[i]);
ItemId idDest = PageGetItemId(page, moveDest[i]);
ItemIdData tmp;
tmp = *idSrc;
*idSrc = *idDest;
*idDest = tmp;
}
spgPageIndexMultiDelete(&bds->spgstate, page,
moveSrc, xlrec.nMove,
SPGIST_PLACEHOLDER, SPGIST_PLACEHOLDER,
InvalidBlockNumber, InvalidOffsetNumber);
for (i = 0; i < xlrec.nChain; i++)
{
SpGistLeafTuple lt;
lt = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, chainSrc[i]));
Assert(lt->tupstate == SPGIST_LIVE);
lt->nextOffset = chainDest[i];
}
MarkBufferDirty(buffer);
if (RelationNeedsWAL(index))
{
XLogRecPtr recptr;
recptr = XLogInsert(RM_SPGIST_ID, XLOG_SPGIST_VACUUM_LEAF, rdata);
PageSetLSN(page, recptr);
}
END_CRIT_SECTION();
}
static void
spgRedoPickSplit(XLogRecPtr lsn, XLogRecord *record)
{
char *ptr = XLogRecGetData(record);
spgxlogPickSplit *xldata = (spgxlogPickSplit *) ptr;
SpGistInnerTuple innerTuple;
SpGistState state;
OffsetNumber *toDelete;
OffsetNumber *toInsert;
uint8 *leafPageSelect;
Buffer srcBuffer;
Buffer destBuffer;
Page page;
int bbi;
int i;
fillFakeState(&state, xldata->stateSrc);
ptr += MAXALIGN(sizeof(spgxlogPickSplit));
innerTuple = (SpGistInnerTuple) ptr;
ptr += innerTuple->size;
toDelete = (OffsetNumber *) ptr;
ptr += MAXALIGN(sizeof(OffsetNumber) * xldata->nDelete);
toInsert = (OffsetNumber *) ptr;
ptr += MAXALIGN(sizeof(OffsetNumber) * xldata->nInsert);
leafPageSelect = (uint8 *) ptr;
ptr += MAXALIGN(sizeof(uint8) * xldata->nInsert);
/* now ptr points to the list of leaf tuples */
/*
* It's a bit tricky to identify which pages have been handled as
* full-page images, so we explicitly count each referenced buffer.
*/
bbi = 0;
if (xldata->blknoSrc == SPGIST_HEAD_BLKNO)
{
/* when splitting root, we touch it only in the guise of new inner */
srcBuffer = InvalidBuffer;
}
else if (xldata->initSrc)
{
/* just re-init the source page */
srcBuffer = XLogReadBuffer(xldata->node, xldata->blknoSrc, true);
Assert(BufferIsValid(srcBuffer));
page = (Page) BufferGetPage(srcBuffer);
SpGistInitBuffer(srcBuffer, SPGIST_LEAF);
/* don't update LSN etc till we're done with it */
}
else
{
/* delete the specified tuples from source page */
if (!(record->xl_info & XLR_SET_BKP_BLOCK(bbi)))
{
srcBuffer = XLogReadBuffer(xldata->node, xldata->blknoSrc, false);
if (BufferIsValid(srcBuffer))
{
page = BufferGetPage(srcBuffer);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
/*
* We have it a bit easier here than in doPickSplit(),
* because we know the inner tuple's location already,
* so we can inject the correct redirection tuple now.
*/
if (!state.isBuild)
spgPageIndexMultiDelete(&state, page,
toDelete, xldata->nDelete,
SPGIST_REDIRECT,
SPGIST_PLACEHOLDER,
xldata->blknoInner,
xldata->offnumInner);
else
spgPageIndexMultiDelete(&state, page,
toDelete, xldata->nDelete,
SPGIST_PLACEHOLDER,
SPGIST_PLACEHOLDER,
InvalidBlockNumber,
InvalidOffsetNumber);
/* don't update LSN etc till we're done with it */
}
}
}
else
srcBuffer = InvalidBuffer;
bbi++;
}
/* try to access dest page if any */
if (xldata->blknoDest == InvalidBlockNumber)
{
destBuffer = InvalidBuffer;
}
else if (xldata->initDest)
{
/* just re-init the dest page */
destBuffer = XLogReadBuffer(xldata->node, xldata->blknoDest, true);
Assert(BufferIsValid(destBuffer));
page = (Page) BufferGetPage(destBuffer);
SpGistInitBuffer(destBuffer, SPGIST_LEAF);
/* don't update LSN etc till we're done with it */
}
else
{
if (!(record->xl_info & XLR_SET_BKP_BLOCK(bbi)))
destBuffer = XLogReadBuffer(xldata->node, xldata->blknoDest, false);
else
destBuffer = InvalidBuffer;
bbi++;
}
/* restore leaf tuples to src and/or dest page */
for (i = 0; i < xldata->nInsert; i++)
{
SpGistLeafTuple lt = (SpGistLeafTuple) ptr;
Buffer leafBuffer;
ptr += lt->size;
leafBuffer = leafPageSelect[i] ? destBuffer : srcBuffer;
if (!BufferIsValid(leafBuffer))
continue; /* no need to touch this page */
page = BufferGetPage(leafBuffer);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
addOrReplaceTuple(page, (Item) lt, lt->size, toInsert[i]);
}
}
/* Now update src and dest page LSNs */
if (BufferIsValid(srcBuffer))
{
page = BufferGetPage(srcBuffer);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
PageSetLSN(page, lsn);
PageSetTLI(page, ThisTimeLineID);
MarkBufferDirty(srcBuffer);
}
UnlockReleaseBuffer(srcBuffer);
}
if (BufferIsValid(destBuffer))
{
page = BufferGetPage(destBuffer);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
PageSetLSN(page, lsn);
PageSetTLI(page, ThisTimeLineID);
MarkBufferDirty(destBuffer);
}
UnlockReleaseBuffer(destBuffer);
}
/* restore new inner tuple */
if (!(record->xl_info & XLR_SET_BKP_BLOCK(bbi)))
{
Buffer buffer = XLogReadBuffer(xldata->node, xldata->blknoInner,
xldata->initInner);
if (BufferIsValid(buffer))
{
page = BufferGetPage(buffer);
if (xldata->initInner)
SpGistInitBuffer(buffer, 0);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
addOrReplaceTuple(page, (Item) innerTuple, innerTuple->size,
xldata->offnumInner);
/* if inner is also parent, update link while we're here */
if (xldata->blknoInner == xldata->blknoParent)
{
SpGistInnerTuple parent;
parent = (SpGistInnerTuple) PageGetItem(page,
PageGetItemId(page, xldata->offnumParent));
spgUpdateNodeLink(parent, xldata->nodeI,
xldata->blknoInner, xldata->offnumInner);
}
PageSetLSN(page, lsn);
PageSetTLI(page, ThisTimeLineID);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
}
bbi++;
/* update parent downlink, unless we did it above */
if (xldata->blknoParent == InvalidBlockNumber)
{
/* no parent cause we split the root */
Assert(xldata->blknoInner == SPGIST_HEAD_BLKNO);
}
else if (xldata->blknoInner != xldata->blknoParent)
{
if (!(record->xl_info & XLR_SET_BKP_BLOCK(bbi)))
{
Buffer buffer = XLogReadBuffer(xldata->node, xldata->blknoParent, false);
if (BufferIsValid(buffer))
{
page = BufferGetPage(buffer);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
SpGistInnerTuple parent;
parent = (SpGistInnerTuple) PageGetItem(page,
PageGetItemId(page, xldata->offnumParent));
spgUpdateNodeLink(parent, xldata->nodeI,
xldata->blknoInner, xldata->offnumInner);
PageSetLSN(page, lsn);
PageSetTLI(page, ThisTimeLineID);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
}
}
}
static void
spgRedoMoveLeafs(XLogRecPtr lsn, XLogRecord *record)
{
char *ptr = XLogRecGetData(record);
spgxlogMoveLeafs *xldata = (spgxlogMoveLeafs *) ptr;
SpGistState state;
OffsetNumber *toDelete;
OffsetNumber *toInsert;
int nInsert;
Buffer buffer;
Page page;
fillFakeState(&state, xldata->stateSrc);
nInsert = xldata->replaceDead ? 1 : xldata->nMoves + 1;
ptr += MAXALIGN(sizeof(spgxlogMoveLeafs));
toDelete = (OffsetNumber *) ptr;
ptr += MAXALIGN(sizeof(OffsetNumber) * xldata->nMoves);
toInsert = (OffsetNumber *) ptr;
ptr += MAXALIGN(sizeof(OffsetNumber) * nInsert);
/* now ptr points to the list of leaf tuples */
/* Insert tuples on the dest page (do first, so redirect is valid) */
if (!(record->xl_info & XLR_BKP_BLOCK_2))
{
buffer = XLogReadBuffer(xldata->node, xldata->blknoDst,
xldata->newPage);
if (BufferIsValid(buffer))
{
page = BufferGetPage(buffer);
if (xldata->newPage)
SpGistInitBuffer(buffer, SPGIST_LEAF);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
int i;
for (i = 0; i < nInsert; i++)
{
SpGistLeafTuple lt = (SpGistLeafTuple) ptr;
addOrReplaceTuple(page, (Item) lt, lt->size, toInsert[i]);
ptr += lt->size;
}
PageSetLSN(page, lsn);
PageSetTLI(page, ThisTimeLineID);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
}
/* Delete tuples from the source page, inserting a redirection pointer */
if (!(record->xl_info & XLR_BKP_BLOCK_1))
{
buffer = XLogReadBuffer(xldata->node, xldata->blknoSrc, false);
if (BufferIsValid(buffer))
{
page = BufferGetPage(buffer);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
spgPageIndexMultiDelete(&state, page, toDelete, xldata->nMoves,
state.isBuild ? SPGIST_PLACEHOLDER : SPGIST_REDIRECT,
SPGIST_PLACEHOLDER,
xldata->blknoDst,
toInsert[nInsert - 1]);
PageSetLSN(page, lsn);
PageSetTLI(page, ThisTimeLineID);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
}
/* And update the parent downlink */
if (!(record->xl_info & XLR_BKP_BLOCK_3))
{
buffer = XLogReadBuffer(xldata->node, xldata->blknoParent, false);
if (BufferIsValid(buffer))
{
page = BufferGetPage(buffer);
if (!XLByteLE(lsn, PageGetLSN(page)))
{
SpGistInnerTuple tuple;
tuple = (SpGistInnerTuple) PageGetItem(page,
PageGetItemId(page, xldata->offnumParent));
spgUpdateNodeLink(tuple, xldata->nodeI,
xldata->blknoDst, toInsert[nInsert - 1]);
PageSetLSN(page, lsn);
PageSetTLI(page, ThisTimeLineID);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
}
}
static void
spgRedoPickSplit(XLogRecPtr lsn, XLogRecord *record)
{
char *ptr = XLogRecGetData(record);
spgxlogPickSplit *xldata = (spgxlogPickSplit *) ptr;
char *innerTuple;
SpGistInnerTupleData innerTupleHdr;
SpGistState state;
OffsetNumber *toDelete;
OffsetNumber *toInsert;
uint8 *leafPageSelect;
Buffer srcBuffer;
Buffer destBuffer;
Page srcPage;
Page destPage;
Buffer innerBuffer;
Page page;
int bbi;
int i;
XLogRedoAction action;
fillFakeState(&state, xldata->stateSrc);
ptr += SizeOfSpgxlogPickSplit;
toDelete = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nDelete;
toInsert = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nInsert;
leafPageSelect = (uint8 *) ptr;
ptr += sizeof(uint8) * xldata->nInsert;
innerTuple = ptr;
/* the inner tuple is unaligned, so make a copy to access its header */
memcpy(&innerTupleHdr, innerTuple, sizeof(SpGistInnerTupleData));
ptr += innerTupleHdr.size;
/* now ptr points to the list of leaf tuples */
/*
* It's a bit tricky to identify which pages have been handled as
* full-page images, so we explicitly count each referenced buffer.
*/
bbi = 0;
if (SpGistBlockIsRoot(xldata->blknoSrc))
{
/* when splitting root, we touch it only in the guise of new inner */
srcBuffer = InvalidBuffer;
srcPage = NULL;
}
else if (xldata->initSrc)
{
/* just re-init the source page */
srcBuffer = XLogReadBuffer(xldata->node, xldata->blknoSrc, true);
Assert(BufferIsValid(srcBuffer));
srcPage = (Page) BufferGetPage(srcBuffer);
SpGistInitBuffer(srcBuffer,
SPGIST_LEAF | (xldata->storesNulls ? SPGIST_NULLS : 0));
/* don't update LSN etc till we're done with it */
}
else
{
/*
* Delete the specified tuples from source page. (In case we're in
* Hot Standby, we need to hold lock on the page till we're done
* inserting leaf tuples and the new inner tuple, else the added
* redirect tuple will be a dangling link.)
*/
if (XLogReadBufferForRedo(lsn, record, bbi,
xldata->node, xldata->blknoSrc,
&srcBuffer) == BLK_NEEDS_REDO)
{
srcPage = BufferGetPage(srcBuffer);
/*
* We have it a bit easier here than in doPickSplit(), because we
* know the inner tuple's location already, so we can inject the
* correct redirection tuple now.
*/
if (!state.isBuild)
spgPageIndexMultiDelete(&state, srcPage,
toDelete, xldata->nDelete,
SPGIST_REDIRECT,
SPGIST_PLACEHOLDER,
xldata->blknoInner,
xldata->offnumInner);
else
spgPageIndexMultiDelete(&state, srcPage,
toDelete, xldata->nDelete,
SPGIST_PLACEHOLDER,
SPGIST_PLACEHOLDER,
InvalidBlockNumber,
InvalidOffsetNumber);
/* don't update LSN etc till we're done with it */
}
else
{
srcPage = NULL; /* don't do any page updates */
}
bbi++;
}
/* try to access dest page if any */
if (xldata->blknoDest == InvalidBlockNumber)
{
destBuffer = InvalidBuffer;
destPage = NULL;
}
else if (xldata->initDest)
{
/* just re-init the dest page */
destBuffer = XLogReadBuffer(xldata->node, xldata->blknoDest, true);
Assert(BufferIsValid(destBuffer));
destPage = (Page) BufferGetPage(destBuffer);
SpGistInitBuffer(destBuffer,
SPGIST_LEAF | (xldata->storesNulls ? SPGIST_NULLS : 0));
/* don't update LSN etc till we're done with it */
}
else
{
/*
* We could probably release the page lock immediately in the
* full-page-image case, but for safety let's hold it till later.
*/
if (XLogReadBufferForRedo(lsn, record, bbi,
xldata->node, xldata->blknoDest,
&destBuffer) == BLK_NEEDS_REDO)
{
destPage = (Page) BufferGetPage(destBuffer);
}
else
{
destPage = NULL; /* don't do any page updates */
}
bbi++;
}
/* restore leaf tuples to src and/or dest page */
for (i = 0; i < xldata->nInsert; i++)
{
char *leafTuple;
SpGistLeafTupleData leafTupleHdr;
/* the tuples are not aligned, so must copy to access the size field. */
leafTuple = ptr;
memcpy(&leafTupleHdr, leafTuple, sizeof(SpGistLeafTupleData));
ptr += leafTupleHdr.size;
page = leafPageSelect[i] ? destPage : srcPage;
if (page == NULL)
continue; /* no need to touch this page */
addOrReplaceTuple(page, (Item) leafTuple, leafTupleHdr.size,
toInsert[i]);
}
/* Now update src and dest page LSNs if needed */
if (srcPage != NULL)
{
PageSetLSN(srcPage, lsn);
MarkBufferDirty(srcBuffer);
}
if (destPage != NULL)
{
PageSetLSN(destPage, lsn);
MarkBufferDirty(destBuffer);
}
/* restore new inner tuple */
if (xldata->initInner)
{
innerBuffer = XLogReadBuffer(xldata->node, xldata->blknoInner, true);
SpGistInitBuffer(innerBuffer,
(xldata->storesNulls ? SPGIST_NULLS : 0));
action = BLK_NEEDS_REDO;
}
else
action = XLogReadBufferForRedo(lsn, record, bbi, xldata->node,
xldata->blknoInner, &innerBuffer);
if (action == BLK_NEEDS_REDO)
{
page = BufferGetPage(innerBuffer);
addOrReplaceTuple(page, (Item) innerTuple, innerTupleHdr.size,
xldata->offnumInner);
/* if inner is also parent, update link while we're here */
if (xldata->blknoInner == xldata->blknoParent)
{
SpGistInnerTuple parent;
parent = (SpGistInnerTuple) PageGetItem(page,
PageGetItemId(page, xldata->offnumParent));
spgUpdateNodeLink(parent, xldata->nodeI,
xldata->blknoInner, xldata->offnumInner);
}
PageSetLSN(page, lsn);
MarkBufferDirty(innerBuffer);
}
if (BufferIsValid(innerBuffer))
UnlockReleaseBuffer(innerBuffer);
bbi++;
/*
* Now we can release the leaf-page locks. It's okay to do this before
* updating the parent downlink.
*/
if (BufferIsValid(srcBuffer))
UnlockReleaseBuffer(srcBuffer);
if (BufferIsValid(destBuffer))
UnlockReleaseBuffer(destBuffer);
/* update parent downlink, unless we did it above */
if (xldata->blknoParent == InvalidBlockNumber)
{
/* no parent cause we split the root */
Assert(SpGistBlockIsRoot(xldata->blknoInner));
}
else if (xldata->blknoInner != xldata->blknoParent)
{
Buffer parentBuffer;
if (XLogReadBufferForRedo(lsn, record, bbi,
xldata->node, xldata->blknoParent,
&parentBuffer) == BLK_NEEDS_REDO)
{
SpGistInnerTuple parent;
page = BufferGetPage(parentBuffer);
parent = (SpGistInnerTuple) PageGetItem(page,
PageGetItemId(page, xldata->offnumParent));
spgUpdateNodeLink(parent, xldata->nodeI,
xldata->blknoInner, xldata->offnumInner);
PageSetLSN(page, lsn);
MarkBufferDirty(parentBuffer);
}
if (BufferIsValid(parentBuffer))
UnlockReleaseBuffer(parentBuffer);
}
}
static void
spgRedoMoveLeafs(XLogRecPtr lsn, XLogRecord *record)
{
char *ptr = XLogRecGetData(record);
spgxlogMoveLeafs *xldata = (spgxlogMoveLeafs *) ptr;
SpGistState state;
OffsetNumber *toDelete;
OffsetNumber *toInsert;
int nInsert;
Buffer buffer;
Page page;
XLogRedoAction action;
fillFakeState(&state, xldata->stateSrc);
nInsert = xldata->replaceDead ? 1 : xldata->nMoves + 1;
ptr += SizeOfSpgxlogMoveLeafs;
toDelete = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nMoves;
toInsert = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * nInsert;
/* now ptr points to the list of leaf tuples */
/*
* In normal operation we would have all three pages (source, dest, and
* parent) locked simultaneously; but in WAL replay it should be safe to
* update them one at a time, as long as we do it in the right order.
*/
/* Insert tuples on the dest page (do first, so redirect is valid) */
if (xldata->newPage)
{
buffer = XLogReadBuffer(xldata->node, xldata->blknoDst, true);
SpGistInitBuffer(buffer,
SPGIST_LEAF | (xldata->storesNulls ? SPGIST_NULLS : 0));
action = BLK_NEEDS_REDO;
}
else
action = XLogReadBufferForRedo(lsn, record, 1,
xldata->node, xldata->blknoDst,
&buffer);
if (action == BLK_NEEDS_REDO)
{
int i;
page = BufferGetPage(buffer);
for (i = 0; i < nInsert; i++)
{
char *leafTuple;
SpGistLeafTupleData leafTupleHdr;
/*
* the tuples are not aligned, so must copy to access the size
* field.
*/
leafTuple = ptr;
memcpy(&leafTupleHdr, leafTuple, sizeof(SpGistLeafTupleData));
addOrReplaceTuple(page, (Item) leafTuple,
leafTupleHdr.size, toInsert[i]);
ptr += leafTupleHdr.size;
}
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
if (BufferIsValid(buffer))
UnlockReleaseBuffer(buffer);
/* Delete tuples from the source page, inserting a redirection pointer */
if (XLogReadBufferForRedo(lsn, record, 0, xldata->node, xldata->blknoSrc,
&buffer) == BLK_NEEDS_REDO)
{
page = BufferGetPage(buffer);
spgPageIndexMultiDelete(&state, page, toDelete, xldata->nMoves,
state.isBuild ? SPGIST_PLACEHOLDER : SPGIST_REDIRECT,
SPGIST_PLACEHOLDER,
xldata->blknoDst,
toInsert[nInsert - 1]);
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
if (BufferIsValid(buffer))
UnlockReleaseBuffer(buffer);
/* And update the parent downlink */
if (XLogReadBufferForRedo(lsn, record, 2, xldata->node, xldata->blknoParent,
&buffer) == BLK_NEEDS_REDO)
{
SpGistInnerTuple tuple;
page = BufferGetPage(buffer);
tuple = (SpGistInnerTuple) PageGetItem(page,
PageGetItemId(page, xldata->offnumParent));
spgUpdateNodeLink(tuple, xldata->nodeI,
xldata->blknoDst, toInsert[nInsert - 1]);
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
if (BufferIsValid(buffer))
UnlockReleaseBuffer(buffer);
}
static void
spgRedoVacuumLeaf(XLogReaderState *record)
{
XLogRecPtr lsn = record->EndRecPtr;
char *ptr = XLogRecGetData(record);
spgxlogVacuumLeaf *xldata = (spgxlogVacuumLeaf *) ptr;
OffsetNumber *toDead;
OffsetNumber *toPlaceholder;
OffsetNumber *moveSrc;
OffsetNumber *moveDest;
OffsetNumber *chainSrc;
OffsetNumber *chainDest;
SpGistState state;
Buffer buffer;
Page page;
int i;
fillFakeState(&state, xldata->stateSrc);
ptr += SizeOfSpgxlogVacuumLeaf;
toDead = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nDead;
toPlaceholder = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nPlaceholder;
moveSrc = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nMove;
moveDest = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nMove;
chainSrc = (OffsetNumber *) ptr;
ptr += sizeof(OffsetNumber) * xldata->nChain;
chainDest = (OffsetNumber *) ptr;
if (XLogReadBufferForRedo(record, 0, &buffer) == BLK_NEEDS_REDO)
{
page = BufferGetPage(buffer);
spgPageIndexMultiDelete(&state, page,
toDead, xldata->nDead,
SPGIST_DEAD, SPGIST_DEAD,
InvalidBlockNumber,
InvalidOffsetNumber);
spgPageIndexMultiDelete(&state, page,
toPlaceholder, xldata->nPlaceholder,
SPGIST_PLACEHOLDER, SPGIST_PLACEHOLDER,
InvalidBlockNumber,
InvalidOffsetNumber);
/* see comments in vacuumLeafPage() */
for (i = 0; i < xldata->nMove; i++)
{
ItemId idSrc = PageGetItemId(page, moveSrc[i]);
ItemId idDest = PageGetItemId(page, moveDest[i]);
ItemIdData tmp;
tmp = *idSrc;
*idSrc = *idDest;
*idDest = tmp;
}
spgPageIndexMultiDelete(&state, page,
moveSrc, xldata->nMove,
SPGIST_PLACEHOLDER, SPGIST_PLACEHOLDER,
InvalidBlockNumber,
InvalidOffsetNumber);
for (i = 0; i < xldata->nChain; i++)
{
SpGistLeafTuple lt;
lt = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, chainSrc[i]));
Assert(lt->tupstate == SPGIST_LIVE);
lt->nextOffset = chainDest[i];
}
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
if (BufferIsValid(buffer))
UnlockReleaseBuffer(buffer);
}
