static void
ginRedoVacuumDataLeafPage(XLogRecPtr lsn, XLogRecord *record)
{
ginxlogVacuumDataLeafPage *xlrec = (ginxlogVacuumDataLeafPage *) XLogRecGetData(record);
Buffer buffer;
Page page;
/* If we have a full-page image, restore it and we're done */
if (record->xl_info & XLR_BKP_BLOCK(0))
{
(void) RestoreBackupBlock(lsn, record, 0, false, false);
return;
}
buffer = XLogReadBuffer(xlrec->node, xlrec->blkno, false);
if (!BufferIsValid(buffer))
return;
page = (Page) BufferGetPage(buffer);
Assert(GinPageIsLeaf(page));
Assert(GinPageIsData(page));
if (lsn > PageGetLSN(page))
{
ginRedoRecompress(page, &xlrec->data);
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
/*
* Replay the clearing of F_FOLLOW_RIGHT flag on a child page.
*
* Even if the WAL record includes a full-page image, we have to update the
* follow-right flag, because that change is not included in the full-page
* image. To be sure that the intermediate state with the wrong flag value is
* not visible to concurrent Hot Standby queries, this function handles
* restoring the full-page image as well as updating the flag. (Note that
* we never need to do anything else to the child page in the current WAL
* action.)
*/
static void
gistRedoClearFollowRight(XLogRecPtr lsn, XLogRecord *record, int block_index,
RelFileNode node, BlockNumber childblkno)
{
Buffer buffer;
Page page;
if (record->xl_info & XLR_BKP_BLOCK(block_index))
buffer = RestoreBackupBlock(lsn, record, block_index, false, true);
else
{
buffer = XLogReadBuffer(node, childblkno, false);
if (!BufferIsValid(buffer))
return; /* page was deleted, nothing to do */
}
page = (Page) BufferGetPage(buffer);
/*
* Note that we still update the page even if page LSN is equal to the LSN
* of this record, because the updated NSN is not included in the full
* page image.
*/
if (lsn >= PageGetLSN(page))
{
GistPageSetNSN(page, lsn);
GistClearFollowRight(page);
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
/*
* XLogReadBufferForRedoExtended
* Like XLogReadBufferForRedo, but with extra options.
*
* If mode is RBM_ZERO or RBM_ZERO_ON_ERROR, if the page doesn't exist, the
* relation is extended with all-zeroes pages up to the referenced block
* number. In RBM_ZERO mode, the return value is always BLK_NEEDS_REDO.
*
* If 'get_cleanup_lock' is true, a "cleanup lock" is acquired on the buffer
* using LockBufferForCleanup(), instead of a regular exclusive lock.
*/
XLogRedoAction
XLogReadBufferForRedoExtended(XLogRecPtr lsn, XLogRecord *record,
int block_index, RelFileNode rnode,
ForkNumber forkno, BlockNumber blkno,
ReadBufferMode mode, bool get_cleanup_lock,
Buffer *buf)
{
if (record->xl_info & XLR_BKP_BLOCK(block_index))
{
*buf = RestoreBackupBlock(lsn, record, block_index,
get_cleanup_lock, true);
return BLK_RESTORED;
}
else
{
*buf = XLogReadBufferExtended(rnode, forkno, blkno, mode);
if (BufferIsValid(*buf))
{
LockBuffer(*buf, BUFFER_LOCK_EXCLUSIVE);
if (lsn <= PageGetLSN(BufferGetPage(*buf)))
return BLK_DONE;
else
return BLK_NEEDS_REDO;
}
else
return BLK_NOTFOUND;
}
}
static void
ginRedoInsertListPage(XLogRecPtr lsn, XLogRecord *record)
{
ginxlogInsertListPage *data = (ginxlogInsertListPage *) XLogRecGetData(record);
Buffer buffer;
Page page;
OffsetNumber l,
off = FirstOffsetNumber;
int i,
tupsize;
IndexTuple tuples = (IndexTuple) (XLogRecGetData(record) + sizeof(ginxlogInsertListPage));
/* If we have a full-page image, restore it and we're done */
if (record->xl_info & XLR_BKP_BLOCK(0))
{
(void) RestoreBackupBlock(lsn, record, 0, false, false);
return;
}
buffer = XLogReadBuffer(data->node, data->blkno, true);
Assert(BufferIsValid(buffer));
page = BufferGetPage(buffer);
GinInitBuffer(buffer, GIN_LIST);
GinPageGetOpaque(page)->rightlink = data->rightlink;
if (data->rightlink == InvalidBlockNumber)
{
/* tail of sublist */
GinPageSetFullRow(page);
GinPageGetOpaque(page)->maxoff = 1;
}
else
{
GinPageGetOpaque(page)->maxoff = 0;
}
for (i = 0; i < data->ntuples; i++)
{
tupsize = IndexTupleSize(tuples);
l = PageAddItem(page, (Item) tuples, tupsize, off, false, false);
if (l == InvalidOffsetNumber)
elog(ERROR, "failed to add item to index page");
tuples = (IndexTuple) (((char *) tuples) + tupsize);
}
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
UnlockReleaseBuffer(buffer);
}
static void
ginRedoUpdateMetapage(XLogRecPtr lsn, XLogRecord *record)
{
ginxlogUpdateMeta *data = (ginxlogUpdateMeta *) XLogRecGetData(record);
Buffer metabuffer;
Page metapage;
Buffer buffer;
/*
* Restore the metapage. This is essentially the same as a full-page
* image, so restore the metapage unconditionally without looking at the
* LSN, to avoid torn page hazards.
*/
metabuffer = XLogReadBuffer(data->node, GIN_METAPAGE_BLKNO, false);
if (!BufferIsValid(metabuffer))
return; /* assume index was deleted, nothing to do */
metapage = BufferGetPage(metabuffer);
memcpy(GinPageGetMeta(metapage), &data->metadata, sizeof(GinMetaPageData));
PageSetLSN(metapage, lsn);
MarkBufferDirty(metabuffer);
if (data->ntuples > 0)
{
/*
* insert into tail page
*/
if (record->xl_info & XLR_BKP_BLOCK(0))
(void) RestoreBackupBlock(lsn, record, 0, false, false);
else
{
buffer = XLogReadBuffer(data->node, data->metadata.tail, false);
if (BufferIsValid(buffer))
{
Page page = BufferGetPage(buffer);
if (lsn > PageGetLSN(page))
{
OffsetNumber l,
off = (PageIsEmpty(page)) ? FirstOffsetNumber :
OffsetNumberNext(PageGetMaxOffsetNumber(page));
int i,
tupsize;
IndexTuple tuples = (IndexTuple) (XLogRecGetData(record) + sizeof(ginxlogUpdateMeta));
for (i = 0; i < data->ntuples; i++)
{
tupsize = IndexTupleSize(tuples);
l = PageAddItem(page, (Item) tuples, tupsize, off, false, false);
if (l == InvalidOffsetNumber)
elog(ERROR, "failed to add item to index page");
tuples = (IndexTuple) (((char *) tuples) + tupsize);
off++;
}
/*
* Increase counter of heap tuples
*/
GinPageGetOpaque(page)->maxoff++;
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
}
}
else if (data->prevTail != InvalidBlockNumber)
{
/*
* New tail
*/
if (record->xl_info & XLR_BKP_BLOCK(0))
(void) RestoreBackupBlock(lsn, record, 0, false, false);
else
{
buffer = XLogReadBuffer(data->node, data->prevTail, false);
if (BufferIsValid(buffer))
{
Page page = BufferGetPage(buffer);
if (lsn > PageGetLSN(page))
{
GinPageGetOpaque(page)->rightlink = data->newRightlink;
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
}
}
UnlockReleaseBuffer(metabuffer);
}
static void
ginRedoDeletePage(XLogRecPtr lsn, XLogRecord *record)
{
ginxlogDeletePage *data = (ginxlogDeletePage *) XLogRecGetData(record);
Buffer dbuffer;
Buffer pbuffer;
Buffer lbuffer;
Page page;
if (record->xl_info & XLR_BKP_BLOCK(0))
dbuffer = RestoreBackupBlock(lsn, record, 0, false, true);
else
{
dbuffer = XLogReadBuffer(data->node, data->blkno, false);
if (BufferIsValid(dbuffer))
{
page = BufferGetPage(dbuffer);
if (lsn > PageGetLSN(page))
{
Assert(GinPageIsData(page));
GinPageGetOpaque(page)->flags = GIN_DELETED;
PageSetLSN(page, lsn);
MarkBufferDirty(dbuffer);
}
}
}
if (record->xl_info & XLR_BKP_BLOCK(1))
pbuffer = RestoreBackupBlock(lsn, record, 1, false, true);
else
{
pbuffer = XLogReadBuffer(data->node, data->parentBlkno, false);
if (BufferIsValid(pbuffer))
{
page = BufferGetPage(pbuffer);
if (lsn > PageGetLSN(page))
{
Assert(GinPageIsData(page));
Assert(!GinPageIsLeaf(page));
GinPageDeletePostingItem(page, data->parentOffset);
PageSetLSN(page, lsn);
MarkBufferDirty(pbuffer);
}
}
}
if (record->xl_info & XLR_BKP_BLOCK(2))
(void) RestoreBackupBlock(lsn, record, 2, false, false);
else if (data->leftBlkno != InvalidBlockNumber)
{
lbuffer = XLogReadBuffer(data->node, data->leftBlkno, false);
if (BufferIsValid(lbuffer))
{
page = BufferGetPage(lbuffer);
if (lsn > PageGetLSN(page))
{
Assert(GinPageIsData(page));
GinPageGetOpaque(page)->rightlink = data->rightLink;
PageSetLSN(page, lsn);
MarkBufferDirty(lbuffer);
}
UnlockReleaseBuffer(lbuffer);
}
}
if (BufferIsValid(pbuffer))
UnlockReleaseBuffer(pbuffer);
if (BufferIsValid(dbuffer))
UnlockReleaseBuffer(dbuffer);
}
static void
ginRedoSplit(XLogRecPtr lsn, XLogRecord *record)
{
ginxlogSplit *data = (ginxlogSplit *) XLogRecGetData(record);
Buffer lbuffer,
rbuffer;
Page lpage,
rpage;
uint32 flags;
uint32 lflags,
rflags;
char *payload;
bool isLeaf = (data->flags & GIN_INSERT_ISLEAF) != 0;
bool isData = (data->flags & GIN_INSERT_ISDATA) != 0;
bool isRoot = (data->flags & GIN_SPLIT_ROOT) != 0;
payload = XLogRecGetData(record) + sizeof(ginxlogSplit);
/*
* First clear incomplete-split flag on child page if this finishes a
* split
*/
if (!isLeaf)
{
if (record->xl_info & XLR_BKP_BLOCK(0))
(void) RestoreBackupBlock(lsn, record, 0, false, false);
else
ginRedoClearIncompleteSplit(lsn, data->node, data->leftChildBlkno);
}
flags = 0;
if (isLeaf)
flags |= GIN_LEAF;
if (isData)
flags |= GIN_DATA;
if (isLeaf && isData)
flags |= GIN_COMPRESSED;
lflags = rflags = flags;
if (!isRoot)
lflags |= GIN_INCOMPLETE_SPLIT;
lbuffer = XLogReadBuffer(data->node, data->lblkno, true);
Assert(BufferIsValid(lbuffer));
lpage = (Page) BufferGetPage(lbuffer);
GinInitBuffer(lbuffer, lflags);
rbuffer = XLogReadBuffer(data->node, data->rblkno, true);
Assert(BufferIsValid(rbuffer));
rpage = (Page) BufferGetPage(rbuffer);
GinInitBuffer(rbuffer, rflags);
GinPageGetOpaque(lpage)->rightlink = BufferGetBlockNumber(rbuffer);
GinPageGetOpaque(rpage)->rightlink = isRoot ? InvalidBlockNumber : data->rrlink;
/* Do the tree-type specific portion to restore the page contents */
if (isData)
ginRedoSplitData(lpage, rpage, payload);
else
ginRedoSplitEntry(lpage, rpage, payload);
PageSetLSN(rpage, lsn);
MarkBufferDirty(rbuffer);
PageSetLSN(lpage, lsn);
MarkBufferDirty(lbuffer);
if (isRoot)
{
BlockNumber rootBlkno = data->rrlink;
Buffer rootBuf = XLogReadBuffer(data->node, rootBlkno, true);
Page rootPage = BufferGetPage(rootBuf);
GinInitBuffer(rootBuf, flags & ~GIN_LEAF & ~GIN_COMPRESSED);
if (isData)
{
Assert(rootBlkno != GIN_ROOT_BLKNO);
ginDataFillRoot(NULL, BufferGetPage(rootBuf),
BufferGetBlockNumber(lbuffer),
BufferGetPage(lbuffer),
BufferGetBlockNumber(rbuffer),
BufferGetPage(rbuffer));
}
else
{
Assert(rootBlkno == GIN_ROOT_BLKNO);
ginEntryFillRoot(NULL, BufferGetPage(rootBuf),
BufferGetBlockNumber(lbuffer),
BufferGetPage(lbuffer),
BufferGetBlockNumber(rbuffer),
BufferGetPage(rbuffer));
}
PageSetLSN(rootPage, lsn);
MarkBufferDirty(rootBuf);
UnlockReleaseBuffer(rootBuf);
}
UnlockReleaseBuffer(rbuffer);
UnlockReleaseBuffer(lbuffer);
}
static void
ginRedoInsert(XLogRecPtr lsn, XLogRecord *record)
{
ginxlogInsert *data = (ginxlogInsert *) XLogRecGetData(record);
Buffer buffer;
Page page;
char *payload;
BlockNumber leftChildBlkno = InvalidBlockNumber;
BlockNumber rightChildBlkno = InvalidBlockNumber;
bool isLeaf = (data->flags & GIN_INSERT_ISLEAF) != 0;
payload = XLogRecGetData(record) + sizeof(ginxlogInsert);
/*
* First clear incomplete-split flag on child page if this finishes a
* split.
*/
if (!isLeaf)
{
leftChildBlkno = BlockIdGetBlockNumber((BlockId) payload);
payload += sizeof(BlockIdData);
rightChildBlkno = BlockIdGetBlockNumber((BlockId) payload);
payload += sizeof(BlockIdData);
if (record->xl_info & XLR_BKP_BLOCK(0))
(void) RestoreBackupBlock(lsn, record, 0, false, false);
else
ginRedoClearIncompleteSplit(lsn, data->node, leftChildBlkno);
}
/* If we have a full-page image, restore it and we're done */
if (record->xl_info & XLR_BKP_BLOCK(isLeaf ? 0 : 1))
{
(void) RestoreBackupBlock(lsn, record, isLeaf ? 0 : 1, false, false);
return;
}
buffer = XLogReadBuffer(data->node, data->blkno, false);
if (!BufferIsValid(buffer))
return; /* page was deleted, nothing to do */
page = (Page) BufferGetPage(buffer);
if (lsn > PageGetLSN(page))
{
/* How to insert the payload is tree-type specific */
if (data->flags & GIN_INSERT_ISDATA)
{
Assert(GinPageIsData(page));
ginRedoInsertData(buffer, isLeaf, rightChildBlkno, payload);
}
else
{
Assert(!GinPageIsData(page));
ginRedoInsertEntry(buffer, isLeaf, rightChildBlkno, payload);
}
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
UnlockReleaseBuffer(buffer);
}
/*
* redo any page update (except page split)
*/
static void
gistRedoPageUpdateRecord(XLogRecPtr lsn, XLogRecord *record)
{
char *begin = XLogRecGetData(record);
gistxlogPageUpdate *xldata = (gistxlogPageUpdate *) begin;
Buffer buffer;
Page page;
char *data;
/*
* We need to acquire and hold lock on target page while updating the left
* child page. If we have a full-page image of target page, getting the
* lock is a side-effect of restoring that image. Note that even if the
* target page no longer exists, we'll still attempt to replay the change
* on the child page.
*/
if (record->xl_info & XLR_BKP_BLOCK(0))
buffer = RestoreBackupBlock(lsn, record, 0, false, true);
else
buffer = XLogReadBuffer(xldata->node, xldata->blkno, false);
/* Fix follow-right data on left child page */
if (BlockNumberIsValid(xldata->leftchild))
gistRedoClearFollowRight(lsn, record, 1,
xldata->node, xldata->leftchild);
/* Done if target page no longer exists */
if (!BufferIsValid(buffer))
return;
/* nothing more to do if page was backed up (and no info to do it with) */
if (record->xl_info & XLR_BKP_BLOCK(0))
{
UnlockReleaseBuffer(buffer);
return;
}
page = (Page) BufferGetPage(buffer);
/* nothing more to do if change already applied */
if (lsn <= PageGetLSN(page))
{
UnlockReleaseBuffer(buffer);
return;
}
data = begin + sizeof(gistxlogPageUpdate);
/* Delete old tuples */
if (xldata->ntodelete > 0)
{
int i;
OffsetNumber *todelete = (OffsetNumber *) data;
data += sizeof(OffsetNumber) * xldata->ntodelete;
for (i = 0; i < xldata->ntodelete; i++)
PageIndexTupleDelete(page, todelete[i]);
if (GistPageIsLeaf(page))
GistMarkTuplesDeleted(page);
}
/* add tuples */
if (data - begin < record->xl_len)
{
OffsetNumber off = (PageIsEmpty(page)) ? FirstOffsetNumber :
OffsetNumberNext(PageGetMaxOffsetNumber(page));
while (data - begin < record->xl_len)
{
IndexTuple itup = (IndexTuple) data;
Size sz = IndexTupleSize(itup);
OffsetNumber l;
data += sz;
l = PageAddItem(page, (Item) itup, sz, off, false, false);
if (l == InvalidOffsetNumber)
elog(ERROR, "failed to add item to GiST index page, size %d bytes",
(int) sz);
off++;
}
}
else
{
/*
* special case: leafpage, nothing to insert, nothing to delete, then
* vacuum marks page
*/
if (GistPageIsLeaf(page) && xldata->ntodelete == 0)
GistClearTuplesDeleted(page);
}
if (!GistPageIsLeaf(page) &&
PageGetMaxOffsetNumber(page) == InvalidOffsetNumber &&
xldata->blkno == GIST_ROOT_BLKNO)
{
/*
* all links on non-leaf root page was deleted by vacuum full, so root
* page becomes a leaf
*/
GistPageSetLeaf(page);
}
GistPageGetOpaque(page)->rightlink = InvalidBlockNumber;
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
UnlockReleaseBuffer(buffer);
}
static void
spgRedoAddNode(XLogRecPtr lsn, XLogRecord *record)
{
char *ptr = XLogRecGetData(record);
spgxlogAddNode *xldata = (spgxlogAddNode *) ptr;
char *innerTuple;
SpGistInnerTupleData innerTupleHdr;
SpGistState state;
Buffer buffer;
Page page;
int bbi;
XLogRedoAction action;
ptr += sizeof(spgxlogAddNode);
innerTuple = ptr;
/* the tuple is unaligned, so make a copy to access its header */
memcpy(&innerTupleHdr, innerTuple, sizeof(SpGistInnerTupleData));
fillFakeState(&state, xldata->stateSrc);
if (xldata->blknoNew == InvalidBlockNumber)
{
/* update in place */
Assert(xldata->blknoParent == InvalidBlockNumber);
if (XLogReadBufferForRedo(lsn, record, 0, xldata->node, xldata->blkno,
&buffer) == BLK_NEEDS_REDO)
{
page = BufferGetPage(buffer);
PageIndexTupleDelete(page, xldata->offnum);
if (PageAddItem(page, (Item) innerTuple, innerTupleHdr.size,
xldata->offnum, false, false) != xldata->offnum)
elog(ERROR, "failed to add item of size %u to SPGiST index page",
innerTupleHdr.size);
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
if (BufferIsValid(buffer))
UnlockReleaseBuffer(buffer);
}
else
{
/*
* 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.
*
* The logic here depends on the assumption that blkno != blknoNew,
* else we can't tell which BKP bit goes with which page, and the LSN
* checks could go wrong too.
*/
Assert(xldata->blkno != xldata->blknoNew);
/* Install new tuple first so redirect is valid */
if (xldata->newPage)
{
buffer = XLogReadBuffer(xldata->node, xldata->blknoNew, true);
/* AddNode is not used for nulls pages */
SpGistInitBuffer(buffer, 0);
action = BLK_NEEDS_REDO;
}
else
action = XLogReadBufferForRedo(lsn, record, 1,
xldata->node, xldata->blknoNew,
&buffer);
if (action == BLK_NEEDS_REDO)
{
page = BufferGetPage(buffer);
addOrReplaceTuple(page, (Item) innerTuple,
innerTupleHdr.size, xldata->offnumNew);
/*
* If parent is in this same page, don't advance LSN; doing so
* would fool us into not applying the parent downlink update
* below. We'll update the LSN when we fix the parent downlink.
*/
if (xldata->blknoParent != xldata->blknoNew)
{
PageSetLSN(page, lsn);
}
MarkBufferDirty(buffer);
}
if (BufferIsValid(buffer))
UnlockReleaseBuffer(buffer);
/* Delete old tuple, replacing it with redirect or placeholder tuple */
if (XLogReadBufferForRedo(lsn, record, 0, xldata->node, xldata->blkno,
&buffer) == BLK_NEEDS_REDO)
{
SpGistDeadTuple dt;
page = BufferGetPage(buffer);
if (state.isBuild)
dt = spgFormDeadTuple(&state, SPGIST_PLACEHOLDER,
InvalidBlockNumber,
InvalidOffsetNumber);
else
dt = spgFormDeadTuple(&state, SPGIST_REDIRECT,
xldata->blknoNew,
xldata->offnumNew);
PageIndexTupleDelete(page, xldata->offnum);
if (PageAddItem(page, (Item) dt, dt->size, xldata->offnum,
false, false) != xldata->offnum)
elog(ERROR, "failed to add item of size %u to SPGiST index page",
dt->size);
if (state.isBuild)
SpGistPageGetOpaque(page)->nPlaceholder++;
else
SpGistPageGetOpaque(page)->nRedirection++;
/*
* If parent is in this same page, don't advance LSN; doing so
* would fool us into not applying the parent downlink update
* below. We'll update the LSN when we fix the parent downlink.
*/
if (xldata->blknoParent != xldata->blkno)
{
PageSetLSN(page, lsn);
}
MarkBufferDirty(buffer);
}
if (BufferIsValid(buffer))
UnlockReleaseBuffer(buffer);
/*
* Update parent downlink. Since parent could be in either of the
* previous two buffers, it's a bit tricky to determine which BKP bit
* applies.
*/
if (xldata->blknoParent == xldata->blkno)
bbi = 0;
else if (xldata->blknoParent == xldata->blknoNew)
bbi = 1;
else
bbi = 2;
if (record->xl_info & XLR_BKP_BLOCK(bbi))
{
if (bbi == 2) /* else we already did it */
(void) RestoreBackupBlock(lsn, record, bbi, false, false);
action = BLK_RESTORED;
buffer = InvalidBuffer;
}
else
{
action = XLogReadBufferForRedo(lsn, record, bbi, xldata->node,
xldata->blknoParent, &buffer);
Assert(action != BLK_RESTORED);
}
if (action == BLK_NEEDS_REDO)
{
SpGistInnerTuple innerTuple;
page = BufferGetPage(buffer);
innerTuple = (SpGistInnerTuple) PageGetItem(page,
PageGetItemId(page, xldata->offnumParent));
spgUpdateNodeLink(innerTuple, xldata->nodeI,
xldata->blknoNew, xldata->offnumNew);
PageSetLSN(page, lsn);
MarkBufferDirty(buffer);
}
if (BufferIsValid(buffer))
UnlockReleaseBuffer(buffer);
}
}
