/*
* _bt_initmetapage() -- Fill a page buffer with a correct metapage image
*/
void
_bt_initmetapage(Page page, BlockNumber rootbknum, uint32 level)
{
BTMetaPageData *metad;
BTPageOpaque metaopaque;
_bt_pageinit(page, BLCKSZ);
metad = BTPageGetMeta(page);
metad->btm_magic = BTREE_MAGIC;
metad->btm_version = BTREE_VERSION;
metad->btm_root = rootbknum;
metad->btm_level = level;
metad->btm_fastroot = rootbknum;
metad->btm_fastlevel = level;
metaopaque = (BTPageOpaque) PageGetSpecialPointer(page);
metaopaque->btpo_flags = BTP_META;
/*
* Set pd_lower just past the end of the metadata. This is not essential
* but it makes the page look compressible to xlog.c.
*/
((PageHeader) page)->pd_lower =
((char *) metad + sizeof(BTMetaPageData)) - (char *) page;
}
/*
* _bt_metapinit() -- Initialize the metadata page of a new btree.
*
* Note: this is actually not used for standard btree index building;
* nbtsort.c prefers not to make the metadata page valid until completion
* of build.
*
* Note: there's no real need for any locking here. Since the transaction
* creating the index hasn't committed yet, no one else can even see the index
* much less be trying to use it. (In a REINDEX-in-place scenario, that's
* not true, but we assume the caller holds sufficient locks on the index.)
*/
void
_bt_metapinit(Relation rel)
{
Buffer buf;
Page pg;
BTMetaPageData *metad;
if (RelationGetNumberOfBlocks(rel) != 0)
elog(ERROR, "cannot initialize non-empty btree index \"%s\"",
RelationGetRelationName(rel));
buf = ReadBuffer(rel, P_NEW);
Assert(BufferGetBlockNumber(buf) == BTREE_METAPAGE);
pg = BufferGetPage(buf);
_bt_initmetapage(pg, P_NONE, 0);
metad = BTPageGetMeta(pg);
/* NO ELOG(ERROR) from here till newmeta op is logged */
START_CRIT_SECTION();
/* XLOG stuff */
if (!rel->rd_istemp)
{
xl_btree_newmeta xlrec;
XLogRecPtr recptr;
XLogRecData rdata[1];
xlrec.node = rel->rd_node;
xlrec.meta.root = metad->btm_root;
xlrec.meta.level = metad->btm_level;
xlrec.meta.fastroot = metad->btm_fastroot;
xlrec.meta.fastlevel = metad->btm_fastlevel;
rdata[0].data = (char *) &xlrec;
rdata[0].len = SizeOfBtreeNewmeta;
rdata[0].buffer = InvalidBuffer;
rdata[0].next = NULL;
recptr = XLogInsert(RM_BTREE_ID,
XLOG_BTREE_NEWMETA,
rdata);
PageSetLSN(pg, recptr);
PageSetTLI(pg, ThisTimeLineID);
}
END_CRIT_SECTION();
WriteBuffer(buf);
}
static void
_bt_restore_meta(XLogReaderState *record, uint8 block_id)
{
XLogRecPtr lsn = record->EndRecPtr;
Buffer metabuf;
Page metapg;
BTMetaPageData *md;
BTPageOpaque pageop;
xl_btree_metadata *xlrec;
char *ptr;
Size len;
metabuf = XLogInitBufferForRedo(record, block_id);
ptr = XLogRecGetBlockData(record, block_id, &len);
Assert(len == sizeof(xl_btree_metadata));
Assert(BufferGetBlockNumber(metabuf) == BTREE_METAPAGE);
xlrec = (xl_btree_metadata *) ptr;
metapg = BufferGetPage(metabuf);
_bt_pageinit(metapg, BufferGetPageSize(metabuf));
md = BTPageGetMeta(metapg);
md->btm_magic = BTREE_MAGIC;
md->btm_version = BTREE_VERSION;
md->btm_root = xlrec->root;
md->btm_level = xlrec->level;
md->btm_fastroot = xlrec->fastroot;
md->btm_fastlevel = xlrec->fastlevel;
pageop = (BTPageOpaque) PageGetSpecialPointer(metapg);
pageop->btpo_flags = BTP_META;
/*
* Set pd_lower just past the end of the metadata. This is not essential
* but it makes the page look compressible to xlog.c.
*/
((PageHeader) metapg)->pd_lower =
((char *) md + sizeof(BTMetaPageData)) - (char *) metapg;
PageSetLSN(metapg, lsn);
MarkBufferDirty(metabuf);
UnlockReleaseBuffer(metabuf);
}
/*
* _bt_gettrueroot() -- Get the true root page of the btree.
*
* This is the same as the BT_READ case of _bt_getroot(), except
* we follow the true-root link not the fast-root link.
*
* By the time we acquire lock on the root page, it might have been split and
* not be the true root anymore. This is okay for the present uses of this
* routine; we only really need to be able to move up at least one tree level
* from whatever non-root page we were at. If we ever do need to lock the
* one true root page, we could loop here, re-reading the metapage on each
* failure. (Note that it wouldn't do to hold the lock on the metapage while
* moving to the root --- that'd deadlock against any concurrent root split.)
*/
Buffer
_bt_gettrueroot(Relation rel)
{
Buffer metabuf;
Page metapg;
BTPageOpaque metaopaque;
Buffer rootbuf;
Page rootpage;
BTPageOpaque rootopaque;
BlockNumber rootblkno;
uint32 rootlevel;
BTMetaPageData *metad;
MIRROREDLOCK_BUFMGR_MUST_ALREADY_BE_HELD;
/*
* We don't try to use cached metapage data here, since (a) this path is
* not performance-critical, and (b) if we are here it suggests our cache
* is out-of-date anyway. In light of point (b), it's probably safest to
* actively flush any cached metapage info.
*/
if (rel->rd_amcache)
pfree(rel->rd_amcache);
rel->rd_amcache = NULL;
metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_READ);
metapg = BufferGetPage(metabuf);
metaopaque = (BTPageOpaque) PageGetSpecialPointer(metapg);
metad = BTPageGetMeta(metapg);
if (!(metaopaque->btpo_flags & BTP_META) ||
metad->btm_magic != BTREE_MAGIC)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("index \"%s\" is not a btree",
RelationGetRelationName(rel))));
if (metad->btm_version != BTREE_VERSION)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("version mismatch in index \"%s\": file version %d, code version %d",
RelationGetRelationName(rel),
metad->btm_version, BTREE_VERSION)));
/* if no root page initialized yet, fail */
if (metad->btm_root == P_NONE)
{
_bt_relbuf(rel, metabuf);
return InvalidBuffer;
}
rootblkno = metad->btm_root;
rootlevel = metad->btm_level;
/*
* We are done with the metapage; arrange to release it via first
* _bt_relandgetbuf call
*/
rootbuf = metabuf;
for (;;)
{
rootbuf = _bt_relandgetbuf(rel, rootbuf, rootblkno, BT_READ);
rootpage = BufferGetPage(rootbuf);
rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage);
if (!P_IGNORE(rootopaque))
break;
/* it's dead, Jim. step right one page */
if (P_RIGHTMOST(rootopaque))
elog(ERROR, "no live root page found in index \"%s\"",
RelationGetRelationName(rel));
rootblkno = rootopaque->btpo_next;
}
/* Note: can't check btpo.level on deleted pages */
if (rootopaque->btpo.level != rootlevel)
elog(ERROR, "root page %u of index \"%s\" has level %u, expected %u",
rootblkno, RelationGetRelationName(rel),
rootopaque->btpo.level, rootlevel);
return rootbuf;
}
static Datum
pgstatindex_impl(Relation rel, FunctionCallInfo fcinfo)
{
Datum result;
BlockNumber nblocks;
BlockNumber blkno;
BTIndexStat indexStat;
BufferAccessStrategy bstrategy = GetAccessStrategy(BAS_BULKREAD);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "relation \"%s\" is not a btree index",
RelationGetRelationName(rel));
/*
* Reject attempts to read non-local temporary relations; we would be
* likely to get wrong data since we have no visibility into the owning
* session's local buffers.
*/
if (RELATION_IS_OTHER_TEMP(rel))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
/*
* Read metapage
*/
{
Buffer buffer = ReadBufferExtended(rel, MAIN_FORKNUM, 0, RBM_NORMAL, bstrategy);
Page page = BufferGetPage(buffer);
BTMetaPageData *metad = BTPageGetMeta(page);
indexStat.version = metad->btm_version;
indexStat.level = metad->btm_level;
indexStat.root_blkno = metad->btm_root;
ReleaseBuffer(buffer);
}
/* -- init counters -- */
indexStat.internal_pages = 0;
indexStat.leaf_pages = 0;
indexStat.empty_pages = 0;
indexStat.deleted_pages = 0;
indexStat.max_avail = 0;
indexStat.free_space = 0;
indexStat.fragments = 0;
/*
* Scan all blocks except the metapage
*/
nblocks = RelationGetNumberOfBlocks(rel);
for (blkno = 1; blkno < nblocks; blkno++)
{
Buffer buffer;
Page page;
BTPageOpaque opaque;
CHECK_FOR_INTERRUPTS();
/* Read and lock buffer */
buffer = ReadBufferExtended(rel, MAIN_FORKNUM, blkno, RBM_NORMAL, bstrategy);
LockBuffer(buffer, BUFFER_LOCK_SHARE);
page = BufferGetPage(buffer);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
/* Determine page type, and update totals */
if (P_ISDELETED(opaque))
indexStat.deleted_pages++;
else if (P_IGNORE(opaque))
indexStat.empty_pages++; /* this is the "half dead" state */
else if (P_ISLEAF(opaque))
{
int max_avail;
max_avail = BLCKSZ - (BLCKSZ - ((PageHeader) page)->pd_special + SizeOfPageHeaderData);
indexStat.max_avail += max_avail;
indexStat.free_space += PageGetFreeSpace(page);
indexStat.leaf_pages++;
/*
* If the next leaf is on an earlier block, it means a
* fragmentation.
*/
if (opaque->btpo_next != P_NONE && opaque->btpo_next < blkno)
indexStat.fragments++;
}
else
indexStat.internal_pages++;
/* Unlock and release buffer */
LockBuffer(buffer, BUFFER_LOCK_UNLOCK);
ReleaseBuffer(buffer);
}
relation_close(rel, AccessShareLock);
/*----------------------------
* Build a result tuple
*----------------------------
*/
{
TupleDesc tupleDesc;
int j;
char *values[10];
HeapTuple tuple;
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &tupleDesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
j = 0;
values[j++] = psprintf("%d", indexStat.version);
values[j++] = psprintf("%d", indexStat.level);
values[j++] = psprintf(INT64_FORMAT,
(1 + /* include the metapage in index_size */
indexStat.leaf_pages +
indexStat.internal_pages +
indexStat.deleted_pages +
indexStat.empty_pages) * BLCKSZ);
values[j++] = psprintf("%u", indexStat.root_blkno);
values[j++] = psprintf(INT64_FORMAT, indexStat.internal_pages);
values[j++] = psprintf(INT64_FORMAT, indexStat.leaf_pages);
values[j++] = psprintf(INT64_FORMAT, indexStat.empty_pages);
values[j++] = psprintf(INT64_FORMAT, indexStat.deleted_pages);
if (indexStat.max_avail > 0)
values[j++] = psprintf("%.2f",
100.0 - (double) indexStat.free_space / (double) indexStat.max_avail * 100.0);
else
values[j++] = pstrdup("NaN");
if (indexStat.leaf_pages > 0)
values[j++] = psprintf("%.2f",
(double) indexStat.fragments / (double) indexStat.leaf_pages * 100.0);
else
values[j++] = pstrdup("NaN");
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = HeapTupleGetDatum(tuple);
}
return result;
}
/**
* @brief Read the left-most leaf page by walking down on index tree structure
* from root node.
*
* Process flow
* -# Open index file and read meta page
* -# Get block number of root page
* -# Read "fast root" page
* -# Read left child page until reaching left-most leaf page
*
* After calling this function, the members of BTReader are the following:
* - smgr : Smgr relation of the existing index file.
* - blkno : block number of left-most leaf page. If there is no leaf page,
* InvalidBlockNumber is set.
* - offnum : InvalidOffsetNumber is set.
* - page : Left-most leaf page, or undefined if no leaf page.
*
* @param reader [in/out] B-Tree index reader
* @return true iff there are some tuples
*/
static bool
BTReaderInit(BTReader *reader, Relation rel)
{
BTPageOpaque metaopaque;
BTMetaPageData *metad;
BTPageOpaque opaque;
BlockNumber blkno;
/*
* HACK: We cannot use smgropen because smgrs returned from it
* will be closed automatically when we assign a new file node.
*
* XXX: It might be better to open the previous relfilenode with
* smgropen *after* RelationSetNewRelfilenode.
*/
memset(&reader->smgr, 0, sizeof(reader->smgr));
#if PG_VERSION_NUM >= 90100
reader->smgr.smgr_rnode.node = rel->rd_node;
reader->smgr.smgr_rnode.backend =
rel->rd_backend == MyBackendId ? MyBackendId : InvalidBackendId;
#else
reader->smgr.smgr_rnode = rel->rd_node;
#endif
reader->smgr.smgr_which = 0; /* md.c */
reader->blkno = InvalidBlockNumber;
reader->offnum = InvalidOffsetNumber;
reader->page = palloc(BLCKSZ);
/*
* Read meta page and check sanity of it.
*
* XXX: It might be better to do REINDEX against corrupted indexes
* instead of raising errors because we've spent long time for data
* loading...
*/
BTReaderReadPage(reader, BTREE_METAPAGE);
metaopaque = (BTPageOpaque) PageGetSpecialPointer(reader->page);
metad = BTPageGetMeta(reader->page);
if (!(metaopaque->btpo_flags & BTP_META) ||
metad->btm_magic != BTREE_MAGIC)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("index \"%s\" is not a reader",
RelationGetRelationName(rel))));
if (metad->btm_version != BTREE_VERSION)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("version mismatch in index \"%s\": file version %d,"
" code version %d",
RelationGetRelationName(rel),
metad->btm_version, BTREE_VERSION)));
if (metad->btm_root == P_NONE)
{
/* No root page; We ignore the index in the subsequent build. */
reader->blkno = InvalidBlockNumber;
return false;
}
/* Go to the fast root page. */
blkno = metad->btm_fastroot;
BTReaderReadPage(reader, blkno);
opaque = (BTPageOpaque) PageGetSpecialPointer(reader->page);
/* Walk down to the left-most leaf page */
while (!P_ISLEAF(opaque))
{
ItemId firstid;
IndexTuple itup;
/* Get the block number of the left child */
firstid = PageGetItemId(reader->page, P_FIRSTDATAKEY(opaque));
itup = (IndexTuple) PageGetItem(reader->page, firstid);
blkno = ItemPointerGetBlockNumber(&(itup->t_tid));
/* Go down to children */
for (;;)
{
BTReaderReadPage(reader, blkno);
opaque = (BTPageOpaque) PageGetSpecialPointer(reader->page);
if (!P_IGNORE(opaque))
break;
if (P_RIGHTMOST(opaque))
{
/* We reach end of the index without any valid leaves. */
reader->blkno = InvalidBlockNumber;
return false;
}
blkno = opaque->btpo_next;
}
}
return true;
}
/* ------------------------------------------------
* bt_metap()
*
* Get a btree's meta-page information
*
* Usage: SELECT * FROM bt_metap('t1_pkey')
* ------------------------------------------------
*/
Datum
bt_metap(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_PP(0);
Datum result;
Relation rel;
RangeVar *relrv;
BTMetaPageData *metad;
TupleDesc tupleDesc;
int j;
char *values[6];
Buffer buffer;
Page page;
HeapTuple tuple;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pageinspect functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "relation \"%s\" is not a btree index",
RelationGetRelationName(rel));
/*
* Reject attempts to read non-local temporary relations; we would be
* likely to get wrong data since we have no visibility into the owning
* session's local buffers.
*/
if (RELATION_IS_OTHER_TEMP(rel))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
buffer = ReadBuffer(rel, 0);
LockBuffer(buffer, BUFFER_LOCK_SHARE);
page = BufferGetPage(buffer);
metad = BTPageGetMeta(page);
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &tupleDesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
j = 0;
values[j++] = psprintf("%d", metad->btm_magic);
values[j++] = psprintf("%d", metad->btm_version);
values[j++] = psprintf("%d", metad->btm_root);
values[j++] = psprintf("%d", metad->btm_level);
values[j++] = psprintf("%d", metad->btm_fastroot);
values[j++] = psprintf("%d", metad->btm_fastlevel);
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = HeapTupleGetDatum(tuple);
UnlockReleaseBuffer(buffer);
relation_close(rel, AccessShareLock);
PG_RETURN_DATUM(result);
}
/* ------------------------------------------------
* bt_metap()
*
* Get a btree meta-page information
*
* Usage: SELECT * FROM bt_metap('t1_pkey')
* ------------------------------------------------
*/
Datum
bt_metap(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_P(0);
Buffer buffer;
Relation rel;
RangeVar *relrv;
Datum result;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "bt_metap() can be used only on b-tree index.");
/*
* Reject attempts to read non-local temporary relations; we would
* be likely to get wrong data since we have no visibility into the
* owning session's local buffers.
*/
if (isOtherTempNamespace(RelationGetNamespace(rel)))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
buffer = ReadBuffer(rel, 0);
{
BTMetaPageData *metad;
TupleDesc tupleDesc;
int j;
char *values[BTMETAP_NCOLUMNS];
HeapTuple tuple;
Page page = BufferGetPage(buffer);
metad = BTPageGetMeta(page);
tupleDesc = RelationNameGetTupleDesc(BTMETAP_TYPE);
j = 0;
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_magic);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_version);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_root);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_level);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_fastroot);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_fastlevel);
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = TupleGetDatum(TupleDescGetSlot(tupleDesc), tuple);
}
ReleaseBuffer(buffer);
relation_close(rel, AccessShareLock);
PG_RETURN_DATUM(result);
}
/* ------------------------------------------------------
* pgstatindex()
*
* Usage: SELECT * FROM pgstatindex('t1_pkey');
* ------------------------------------------------------
*/
Datum
pgstatindex(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_P(0);
Relation rel;
RangeVar *relrv;
Datum result;
uint32 nblocks;
uint32 blkno;
BTIndexStat indexStat;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "pgstatindex() can be used only on b-tree index.");
/*-------------------
* Read a metapage
*-------------------
*/
{
Buffer buffer = ReadBuffer(rel, 0);
Page page = BufferGetPage(buffer);
BTMetaPageData *metad = BTPageGetMeta(page);
indexStat.magic = metad->btm_magic;
indexStat.version = metad->btm_version;
indexStat.root_blkno = metad->btm_root;
indexStat.level = metad->btm_level;
indexStat.fastroot = metad->btm_fastroot;
indexStat.fastlevel = metad->btm_fastlevel;
ReleaseBuffer(buffer);
}
nblocks = RelationGetNumberOfBlocks(rel);
/* -- init stat -- */
indexStat.fragments = 0;
indexStat.root_pages = 0;
indexStat.leaf_pages = 0;
indexStat.internal_pages = 0;
indexStat.empty_pages = 0;
indexStat.deleted_pages = 0;
indexStat.max_avail = 0;
indexStat.free_space = 0;
/*-----------------------
* Scan all blocks
*-----------------------
*/
for (blkno = 1; blkno < nblocks; blkno++)
{
Buffer buffer = ReadBuffer(rel, blkno);
BTPageStat stat;
/* scan one page */
stat.blkno = blkno;
GetBTPageStatistics(blkno, buffer, &stat);
/*---------------------
* page status (type)
*---------------------
*/
switch (stat.type)
{
case 'd':
indexStat.deleted_pages++;
break;
case 'l':
indexStat.leaf_pages++;
break;
case 'i':
indexStat.internal_pages++;
break;
case 'e':
indexStat.empty_pages++;
break;
case 'r':
indexStat.root_pages++;
break;
default:
elog(ERROR, "unknown page status.");
}
/* -- leaf fragmentation -- */
indexStat.fragments += stat.fragments;
if (stat.type == 'l')
{
indexStat.max_avail += stat.max_avail;
indexStat.free_space += stat.free_size;
}
ReleaseBuffer(buffer);
}
relation_close(rel, AccessShareLock);
/*----------------------------
* Build a result tuple
*----------------------------
*/
{
TupleDesc tupleDesc;
int j;
char *values[PGSTATINDEX_NCOLUMNS];
HeapTupleData tupleData;
HeapTuple tuple = &tupleData;
tupleDesc = RelationNameGetTupleDesc(PGSTATINDEX_TYPE);
j = 0;
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.version);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.level);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", (indexStat.root_pages +
indexStat.leaf_pages +
indexStat.internal_pages +
indexStat.deleted_pages +
indexStat.empty_pages) * BLCKSZ);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.root_blkno);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.internal_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.leaf_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.empty_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.deleted_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, "%.2f", 100.0 - (float) indexStat.free_space / (float) indexStat.max_avail * 100.0);
values[j] = palloc(32);
snprintf(values[j++], 32, "%.2f", (float) indexStat.fragments / (float) indexStat.leaf_pages * 100.0);
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = TupleGetDatum(TupleDescGetSlot(tupleDesc), tuple);
}
PG_RETURN_DATUM(result);
}
/*
* _bt_pagedel() -- Delete a page from the b-tree.
*
* This action unlinks the page from the b-tree structure, removing all
* pointers leading to it --- but not touching its own left and right links.
* The page cannot be physically reclaimed right away, since other processes
* may currently be trying to follow links leading to the page; they have to
* be allowed to use its right-link to recover. See nbtree/README.
*
* On entry, the target buffer must be pinned and read-locked. This lock and
* pin will be dropped before exiting.
*
* Returns the number of pages successfully deleted (zero on failure; could
* be more than one if parent blocks were deleted).
*
* NOTE: this leaks memory. Rather than trying to clean up everything
* carefully, it's better to run it in a temp context that can be reset
* frequently.
*/
int
_bt_pagedel(Relation rel, Buffer buf, bool vacuum_full)
{
BlockNumber target,
leftsib,
rightsib,
parent;
OffsetNumber poffset,
maxoff;
uint32 targetlevel,
ilevel;
ItemId itemid;
BTItem targetkey,
btitem;
ScanKey itup_scankey;
BTStack stack;
Buffer lbuf,
rbuf,
pbuf;
bool parent_half_dead;
bool parent_one_child;
bool rightsib_empty;
Buffer metabuf = InvalidBuffer;
Page metapg = NULL;
BTMetaPageData *metad = NULL;
Page page;
BTPageOpaque opaque;
/*
* We can never delete rightmost pages nor root pages. While at it, check
* that page is not already deleted and is empty.
*/
page = BufferGetPage(buf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
if (P_RIGHTMOST(opaque) || P_ISROOT(opaque) || P_ISDELETED(opaque) ||
P_FIRSTDATAKEY(opaque) <= PageGetMaxOffsetNumber(page))
{
_bt_relbuf(rel, buf);
return 0;
}
/*
* Save info about page, including a copy of its high key (it must have
* one, being non-rightmost).
*/
target = BufferGetBlockNumber(buf);
targetlevel = opaque->btpo.level;
leftsib = opaque->btpo_prev;
itemid = PageGetItemId(page, P_HIKEY);
targetkey = CopyBTItem((BTItem) PageGetItem(page, itemid));
/*
* We need to get an approximate pointer to the page's parent page. Use
* the standard search mechanism to search for the page's high key; this
* will give us a link to either the current parent or someplace to its
* left (if there are multiple equal high keys). To avoid deadlocks, we'd
* better drop the target page lock first.
*/
_bt_relbuf(rel, buf);
/* we need a scan key to do our search, so build one */
itup_scankey = _bt_mkscankey(rel, &(targetkey->bti_itup));
/* find the leftmost leaf page containing this key */
stack = _bt_search(rel, rel->rd_rel->relnatts, itup_scankey, false,
&lbuf, BT_READ);
/* don't need a pin on that either */
_bt_relbuf(rel, lbuf);
/*
* If we are trying to delete an interior page, _bt_search did more than
* we needed. Locate the stack item pointing to our parent level.
*/
ilevel = 0;
for (;;)
{
if (stack == NULL)
elog(ERROR, "not enough stack items");
if (ilevel == targetlevel)
break;
stack = stack->bts_parent;
ilevel++;
}
/*
* We have to lock the pages we need to modify in the standard order:
* moving right, then up. Else we will deadlock against other writers.
*
* So, we need to find and write-lock the current left sibling of the
* target page. The sibling that was current a moment ago could have
* split, so we may have to move right. This search could fail if either
* the sibling or the target page was deleted by someone else meanwhile;
* if so, give up. (Right now, that should never happen, since page
* deletion is only done in VACUUM and there shouldn't be multiple VACUUMs
* concurrently on the same table.)
*/
if (leftsib != P_NONE)
{
lbuf = _bt_getbuf(rel, leftsib, BT_WRITE);
page = BufferGetPage(lbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
while (P_ISDELETED(opaque) || opaque->btpo_next != target)
{
/* step right one page */
leftsib = opaque->btpo_next;
_bt_relbuf(rel, lbuf);
if (leftsib == P_NONE)
{
elog(LOG, "no left sibling (concurrent deletion?) in \"%s\"",
RelationGetRelationName(rel));
return 0;
}
lbuf = _bt_getbuf(rel, leftsib, BT_WRITE);
page = BufferGetPage(lbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
}
}
else
lbuf = InvalidBuffer;
/*
* Next write-lock the target page itself. It should be okay to take just
* a write lock not a superexclusive lock, since no scans would stop on an
* empty page.
*/
buf = _bt_getbuf(rel, target, BT_WRITE);
page = BufferGetPage(buf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
/*
* Check page is still empty etc, else abandon deletion. The empty check
* is necessary since someone else might have inserted into it while we
* didn't have it locked; the others are just for paranoia's sake.
*/
if (P_RIGHTMOST(opaque) || P_ISROOT(opaque) || P_ISDELETED(opaque) ||
P_FIRSTDATAKEY(opaque) <= PageGetMaxOffsetNumber(page))
{
_bt_relbuf(rel, buf);
if (BufferIsValid(lbuf))
_bt_relbuf(rel, lbuf);
return 0;
}
if (opaque->btpo_prev != leftsib)
elog(ERROR, "left link changed unexpectedly in block %u of \"%s\"",
target, RelationGetRelationName(rel));
/*
* And next write-lock the (current) right sibling.
*/
rightsib = opaque->btpo_next;
rbuf = _bt_getbuf(rel, rightsib, BT_WRITE);
/*
* Next find and write-lock the current parent of the target page. This is
* essentially the same as the corresponding step of splitting. However,
* it's possible for the search to fail (for reasons explained in README).
* If that happens, we recover by searching the whole parent level, which
* is a tad inefficient but doesn't happen often enough to be a problem.
*/
ItemPointerSet(&(stack->bts_btitem.bti_itup.t_tid),
target, P_HIKEY);
pbuf = _bt_getstackbuf(rel, stack, BT_WRITE);
if (pbuf == InvalidBuffer)
{
/* Find the leftmost page in the parent level */
pbuf = _bt_get_endpoint(rel, opaque->btpo.level + 1, false);
stack->bts_blkno = BufferGetBlockNumber(pbuf);
stack->bts_offset = InvalidOffsetNumber;
_bt_relbuf(rel, pbuf);
/* and repeat search from there */
pbuf = _bt_getstackbuf(rel, stack, BT_WRITE);
if (pbuf == InvalidBuffer)
elog(ERROR, "failed to re-find parent key in \"%s\" for deletion target page %u",
RelationGetRelationName(rel), target);
}
parent = stack->bts_blkno;
poffset = stack->bts_offset;
/*
* If the target is the rightmost child of its parent, then we can't
* delete, unless it's also the only child --- in which case the parent
* changes to half-dead status.
*/
page = BufferGetPage(pbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
maxoff = PageGetMaxOffsetNumber(page);
parent_half_dead = false;
parent_one_child = false;
if (poffset >= maxoff)
{
if (poffset == P_FIRSTDATAKEY(opaque))
parent_half_dead = true;
else
{
_bt_relbuf(rel, pbuf);
_bt_relbuf(rel, rbuf);
_bt_relbuf(rel, buf);
if (BufferIsValid(lbuf))
_bt_relbuf(rel, lbuf);
return 0;
}
}
else
{
/* Will there be exactly one child left in this parent? */
if (OffsetNumberNext(P_FIRSTDATAKEY(opaque)) == maxoff)
parent_one_child = true;
}
/*
* If we are deleting the next-to-last page on the target's level, then
* the rightsib is a candidate to become the new fast root. (In theory, it
* might be possible to push the fast root even further down, but the odds
* of doing so are slim, and the locking considerations daunting.)
*
* We can safely acquire a lock on the metapage here --- see comments for
* _bt_newroot().
*/
if (leftsib == P_NONE)
{
page = BufferGetPage(rbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
Assert(opaque->btpo.level == targetlevel);
if (P_RIGHTMOST(opaque))
{
/* rightsib will be the only one left on the level */
metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_WRITE);
metapg = BufferGetPage(metabuf);
metad = BTPageGetMeta(metapg);
/*
* The expected case here is btm_fastlevel == targetlevel+1; if
* the fastlevel is <= targetlevel, something is wrong, and we
* choose to overwrite it to fix it.
*/
if (metad->btm_fastlevel > targetlevel + 1)
{
/* no update wanted */
_bt_relbuf(rel, metabuf);
metabuf = InvalidBuffer;
}
}
}
/*
* Here we begin doing the deletion.
*/
/* No ereport(ERROR) until changes are logged */
START_CRIT_SECTION();
/*
* Update parent. The normal case is a tad tricky because we want to
* delete the target's downlink and the *following* key. Easiest way is
* to copy the right sibling's downlink over the target downlink, and then
* delete the following item.
*/
page = BufferGetPage(pbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
if (parent_half_dead)
{
PageIndexTupleDelete(page, poffset);
opaque->btpo_flags |= BTP_HALF_DEAD;
}
else
{
OffsetNumber nextoffset;
itemid = PageGetItemId(page, poffset);
btitem = (BTItem) PageGetItem(page, itemid);
Assert(ItemPointerGetBlockNumber(&(btitem->bti_itup.t_tid)) == target);
ItemPointerSet(&(btitem->bti_itup.t_tid), rightsib, P_HIKEY);
nextoffset = OffsetNumberNext(poffset);
/* This part is just for double-checking */
itemid = PageGetItemId(page, nextoffset);
btitem = (BTItem) PageGetItem(page, itemid);
if (ItemPointerGetBlockNumber(&(btitem->bti_itup.t_tid)) != rightsib)
elog(PANIC, "right sibling is not next child in \"%s\"",
RelationGetRelationName(rel));
PageIndexTupleDelete(page, nextoffset);
}
/*
* Update siblings' side-links. Note the target page's side-links will
* continue to point to the siblings.
*/
if (BufferIsValid(lbuf))
{
page = BufferGetPage(lbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
Assert(opaque->btpo_next == target);
opaque->btpo_next = rightsib;
}
page = BufferGetPage(rbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
Assert(opaque->btpo_prev == target);
opaque->btpo_prev = leftsib;
rightsib_empty = (P_FIRSTDATAKEY(opaque) > PageGetMaxOffsetNumber(page));
/*
* Mark the page itself deleted. It can be recycled when all current
* transactions are gone; or immediately if we're doing VACUUM FULL.
*/
page = BufferGetPage(buf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
opaque->btpo_flags |= BTP_DELETED;
opaque->btpo.xact =
vacuum_full ? FrozenTransactionId : ReadNewTransactionId();
/* And update the metapage, if needed */
if (BufferIsValid(metabuf))
{
metad->btm_fastroot = rightsib;
metad->btm_fastlevel = targetlevel;
}
/* XLOG stuff */
if (!rel->rd_istemp)
{
xl_btree_delete_page xlrec;
xl_btree_metadata xlmeta;
uint8 xlinfo;
XLogRecPtr recptr;
XLogRecData rdata[5];
XLogRecData *nextrdata;
xlrec.target.node = rel->rd_node;
ItemPointerSet(&(xlrec.target.tid), parent, poffset);
xlrec.deadblk = target;
xlrec.leftblk = leftsib;
xlrec.rightblk = rightsib;
rdata[0].data = (char *) &xlrec;
rdata[0].len = SizeOfBtreeDeletePage;
rdata[0].buffer = InvalidBuffer;
rdata[0].next = nextrdata = &(rdata[1]);
if (BufferIsValid(metabuf))
{
xlmeta.root = metad->btm_root;
xlmeta.level = metad->btm_level;
xlmeta.fastroot = metad->btm_fastroot;
xlmeta.fastlevel = metad->btm_fastlevel;
nextrdata->data = (char *) &xlmeta;
nextrdata->len = sizeof(xl_btree_metadata);
nextrdata->buffer = InvalidBuffer;
nextrdata->next = nextrdata + 1;
nextrdata++;
xlinfo = XLOG_BTREE_DELETE_PAGE_META;
}
else
xlinfo = XLOG_BTREE_DELETE_PAGE;
nextrdata->data = NULL;
nextrdata->len = 0;
nextrdata->next = nextrdata + 1;
nextrdata->buffer = pbuf;
nextrdata->buffer_std = true;
nextrdata++;
nextrdata->data = NULL;
nextrdata->len = 0;
nextrdata->buffer = rbuf;
nextrdata->buffer_std = true;
nextrdata->next = NULL;
if (BufferIsValid(lbuf))
{
nextrdata->next = nextrdata + 1;
nextrdata++;
nextrdata->data = NULL;
nextrdata->len = 0;
nextrdata->buffer = lbuf;
nextrdata->buffer_std = true;
nextrdata->next = NULL;
}
recptr = XLogInsert(RM_BTREE_ID, xlinfo, rdata);
if (BufferIsValid(metabuf))
{
PageSetLSN(metapg, recptr);
PageSetTLI(metapg, ThisTimeLineID);
}
page = BufferGetPage(pbuf);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
page = BufferGetPage(rbuf);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
page = BufferGetPage(buf);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
if (BufferIsValid(lbuf))
{
page = BufferGetPage(lbuf);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
}
}
END_CRIT_SECTION();
/* Write and release buffers */
if (BufferIsValid(metabuf))
_bt_wrtbuf(rel, metabuf);
_bt_wrtbuf(rel, pbuf);
_bt_wrtbuf(rel, rbuf);
_bt_wrtbuf(rel, buf);
if (BufferIsValid(lbuf))
_bt_wrtbuf(rel, lbuf);
/*
* If parent became half dead, recurse to try to delete it. Otherwise, if
* right sibling is empty and is now the last child of the parent, recurse
* to try to delete it. (These cases cannot apply at the same time,
* though the second case might itself recurse to the first.)
*/
if (parent_half_dead)
{
buf = _bt_getbuf(rel, parent, BT_READ);
return _bt_pagedel(rel, buf, vacuum_full) + 1;
}
if (parent_one_child && rightsib_empty)
{
buf = _bt_getbuf(rel, rightsib, BT_READ);
return _bt_pagedel(rel, buf, vacuum_full) + 1;
}
return 1;
}
/*
* _bt_gettrueroot() -- Get the true root page of the btree.
*
* This is the same as the BT_READ case of _bt_getroot(), except
* we follow the true-root link not the fast-root link.
*
* By the time we acquire lock on the root page, it might have been split and
* not be the true root anymore. This is okay for the present uses of this
* routine; we only really need to be able to move up at least one tree level
* from whatever non-root page we were at. If we ever do need to lock the
* one true root page, we could loop here, re-reading the metapage on each
* failure. (Note that it wouldn't do to hold the lock on the metapage while
* moving to the root --- that'd deadlock against any concurrent root split.)
*/
Buffer
_bt_gettrueroot(Relation rel)
{
Buffer metabuf;
Page metapg;
BTPageOpaque metaopaque;
Buffer rootbuf;
Page rootpage;
BTPageOpaque rootopaque;
BlockNumber rootblkno;
uint32 rootlevel;
BTMetaPageData *metad;
metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_READ);
metapg = BufferGetPage(metabuf);
metaopaque = (BTPageOpaque) PageGetSpecialPointer(metapg);
metad = BTPageGetMeta(metapg);
if (!(metaopaque->btpo_flags & BTP_META) ||
metad->btm_magic != BTREE_MAGIC)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("index \"%s\" is not a btree",
RelationGetRelationName(rel))));
if (metad->btm_version != BTREE_VERSION)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("version mismatch in index \"%s\": file version %d, code version %d",
RelationGetRelationName(rel),
metad->btm_version, BTREE_VERSION)));
/* if no root page initialized yet, fail */
if (metad->btm_root == P_NONE)
{
_bt_relbuf(rel, metabuf);
return InvalidBuffer;
}
rootblkno = metad->btm_root;
rootlevel = metad->btm_level;
/*
* We are done with the metapage; arrange to release it via first
* _bt_relandgetbuf call
*/
rootbuf = metabuf;
for (;;)
{
rootbuf = _bt_relandgetbuf(rel, rootbuf, rootblkno, BT_READ);
rootpage = BufferGetPage(rootbuf);
rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage);
if (!P_IGNORE(rootopaque))
break;
/* it's dead, Jim. step right one page */
if (P_RIGHTMOST(rootopaque))
elog(ERROR, "no live root page found in \"%s\"",
RelationGetRelationName(rel));
rootblkno = rootopaque->btpo_next;
}
/* Note: can't check btpo.level on deleted pages */
if (rootopaque->btpo.level != rootlevel)
elog(ERROR, "root page %u of \"%s\" has level %u, expected %u",
rootblkno, RelationGetRelationName(rel),
rootopaque->btpo.level, rootlevel);
return rootbuf;
}
/*
* _bt_getroot() -- Get the root page of the btree.
*
* Since the root page can move around the btree file, we have to read
* its location from the metadata page, and then read the root page
* itself. If no root page exists yet, we have to create one. The
* standard class of race conditions exists here; I think I covered
* them all in the Hopi Indian rain dance of lock requests below.
*
* The access type parameter (BT_READ or BT_WRITE) controls whether
* a new root page will be created or not. If access = BT_READ,
* and no root page exists, we just return InvalidBuffer. For
* BT_WRITE, we try to create the root page if it doesn't exist.
* NOTE that the returned root page will have only a read lock set
* on it even if access = BT_WRITE!
*
* The returned page is not necessarily the true root --- it could be
* a "fast root" (a page that is alone in its level due to deletions).
* Also, if the root page is split while we are "in flight" to it,
* what we will return is the old root, which is now just the leftmost
* page on a probably-not-very-wide level. For most purposes this is
* as good as or better than the true root, so we do not bother to
* insist on finding the true root. We do, however, guarantee to
* return a live (not deleted or half-dead) page.
*
* On successful return, the root page is pinned and read-locked.
* The metadata page is not locked or pinned on exit.
*/
Buffer
_bt_getroot(Relation rel, int access)
{
Buffer metabuf;
Page metapg;
BTPageOpaque metaopaque;
Buffer rootbuf;
Page rootpage;
BTPageOpaque rootopaque;
BlockNumber rootblkno;
uint32 rootlevel;
BTMetaPageData *metad;
metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_READ);
metapg = BufferGetPage(metabuf);
metaopaque = (BTPageOpaque) PageGetSpecialPointer(metapg);
metad = BTPageGetMeta(metapg);
/* sanity-check the metapage */
if (!(metaopaque->btpo_flags & BTP_META) ||
metad->btm_magic != BTREE_MAGIC)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("index \"%s\" is not a btree",
RelationGetRelationName(rel))));
if (metad->btm_version != BTREE_VERSION)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("version mismatch in index \"%s\": file version %d, code version %d",
RelationGetRelationName(rel),
metad->btm_version, BTREE_VERSION)));
/* if no root page initialized yet, do it */
if (metad->btm_root == P_NONE)
{
/* If access = BT_READ, caller doesn't want us to create root yet */
if (access == BT_READ)
{
_bt_relbuf(rel, metabuf);
return InvalidBuffer;
}
/* trade in our read lock for a write lock */
LockBuffer(metabuf, BUFFER_LOCK_UNLOCK);
LockBuffer(metabuf, BT_WRITE);
/*
* Race condition: if someone else initialized the metadata between
* the time we released the read lock and acquired the write lock, we
* must avoid doing it again.
*/
if (metad->btm_root != P_NONE)
{
/*
* Metadata initialized by someone else. In order to guarantee no
* deadlocks, we have to release the metadata page and start all
* over again. (Is that really true? But it's hardly worth trying
* to optimize this case.)
*/
_bt_relbuf(rel, metabuf);
return _bt_getroot(rel, access);
}
/*
* Get, initialize, write, and leave a lock of the appropriate type on
* the new root page. Since this is the first page in the tree, it's
* a leaf as well as the root.
*/
rootbuf = _bt_getbuf(rel, P_NEW, BT_WRITE);
rootblkno = BufferGetBlockNumber(rootbuf);
rootpage = BufferGetPage(rootbuf);
_bt_pageinit(rootpage, BufferGetPageSize(rootbuf));
rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage);
rootopaque->btpo_prev = rootopaque->btpo_next = P_NONE;
rootopaque->btpo_flags = (BTP_LEAF | BTP_ROOT);
rootopaque->btpo.level = 0;
/* NO ELOG(ERROR) till meta is updated */
START_CRIT_SECTION();
metad->btm_root = rootblkno;
metad->btm_level = 0;
metad->btm_fastroot = rootblkno;
metad->btm_fastlevel = 0;
/* XLOG stuff */
if (!rel->rd_istemp)
{
xl_btree_newroot xlrec;
XLogRecPtr recptr;
XLogRecData rdata;
xlrec.node = rel->rd_node;
xlrec.rootblk = rootblkno;
xlrec.level = 0;
rdata.data = (char *) &xlrec;
rdata.len = SizeOfBtreeNewroot;
rdata.buffer = InvalidBuffer;
rdata.next = NULL;
recptr = XLogInsert(RM_BTREE_ID, XLOG_BTREE_NEWROOT, &rdata);
PageSetLSN(rootpage, recptr);
PageSetTLI(rootpage, ThisTimeLineID);
PageSetLSN(metapg, recptr);
PageSetTLI(metapg, ThisTimeLineID);
}
END_CRIT_SECTION();
_bt_wrtnorelbuf(rel, rootbuf);
/*
* swap root write lock for read lock. There is no danger of anyone
* else accessing the new root page while it's unlocked, since no one
* else knows where it is yet.
*/
LockBuffer(rootbuf, BUFFER_LOCK_UNLOCK);
LockBuffer(rootbuf, BT_READ);
/* okay, metadata is correct, write and release it */
_bt_wrtbuf(rel, metabuf);
}
else
{
rootblkno = metad->btm_fastroot;
Assert(rootblkno != P_NONE);
rootlevel = metad->btm_fastlevel;
/*
* We are done with the metapage; arrange to release it via first
* _bt_relandgetbuf call
*/
rootbuf = metabuf;
for (;;)
{
rootbuf = _bt_relandgetbuf(rel, rootbuf, rootblkno, BT_READ);
rootpage = BufferGetPage(rootbuf);
rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage);
if (!P_IGNORE(rootopaque))
break;
/* it's dead, Jim. step right one page */
if (P_RIGHTMOST(rootopaque))
elog(ERROR, "no live root page found in \"%s\"",
RelationGetRelationName(rel));
rootblkno = rootopaque->btpo_next;
}
/* Note: can't check btpo.level on deleted pages */
if (rootopaque->btpo.level != rootlevel)
elog(ERROR, "root page %u of \"%s\" has level %u, expected %u",
rootblkno, RelationGetRelationName(rel),
rootopaque->btpo.level, rootlevel);
}
/*
* By here, we have a pin and read lock on the root page, and no lock set
* on the metadata page. Return the root page's buffer.
*/
return rootbuf;
}
/* ------------------------------------------------------
* pgstatindex()
*
* Usage: SELECT * FROM pgstatindex('t1_pkey');
* ------------------------------------------------------
*/
Datum
pgstatindex(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_P(0);
Relation rel;
RangeVar *relrv;
Datum result;
BlockNumber nblocks;
BlockNumber blkno;
BTIndexStat indexStat;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "relation \"%s\" is not a btree index",
RelationGetRelationName(rel));
/*
* Reject attempts to read non-local temporary relations; we would be
* likely to get wrong data since we have no visibility into the owning
* session's local buffers.
*/
if (RELATION_IS_OTHER_TEMP(rel))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
/*
* Read metapage
*/
{
Buffer buffer = ReadBuffer(rel, 0);
Page page = BufferGetPage(buffer);
BTMetaPageData *metad = BTPageGetMeta(page);
indexStat.version = metad->btm_version;
indexStat.level = metad->btm_level;
indexStat.root_blkno = metad->btm_root;
ReleaseBuffer(buffer);
}
/* -- init counters -- */
indexStat.root_pages = 0;
indexStat.internal_pages = 0;
indexStat.leaf_pages = 0;
indexStat.empty_pages = 0;
indexStat.deleted_pages = 0;
indexStat.max_avail = 0;
indexStat.free_space = 0;
indexStat.fragments = 0;
/*
* Scan all blocks except the metapage
*/
nblocks = RelationGetNumberOfBlocks(rel);
for (blkno = 1; blkno < nblocks; blkno++)
{
Buffer buffer;
Page page;
BTPageOpaque opaque;
/* Read and lock buffer */
buffer = ReadBuffer(rel, blkno);
LockBuffer(buffer, BUFFER_LOCK_SHARE);
page = BufferGetPage(buffer);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
/* Determine page type, and update totals */
if (P_ISLEAF(opaque))
{
int max_avail;
max_avail = BLCKSZ - (BLCKSZ - ((PageHeader) page)->pd_special + SizeOfPageHeaderData);
indexStat.max_avail += max_avail;
indexStat.free_space += PageGetFreeSpace(page);
indexStat.leaf_pages++;
/*
* If the next leaf is on an earlier block, it means a
* fragmentation.
*/
if (opaque->btpo_next != P_NONE && opaque->btpo_next < blkno)
indexStat.fragments++;
}
else if (P_ISDELETED(opaque))
indexStat.deleted_pages++;
else if (P_IGNORE(opaque))
indexStat.empty_pages++;
else if (P_ISROOT(opaque))
indexStat.root_pages++;
else
indexStat.internal_pages++;
/* Unlock and release buffer */
LockBuffer(buffer, BUFFER_LOCK_UNLOCK);
ReleaseBuffer(buffer);
}
relation_close(rel, AccessShareLock);
/*----------------------------
* Build a result tuple
*----------------------------
*/
{
TupleDesc tupleDesc;
int j;
char *values[10];
HeapTuple tuple;
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &tupleDesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
j = 0;
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.version);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.level);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT,
(indexStat.root_pages +
indexStat.leaf_pages +
indexStat.internal_pages +
indexStat.deleted_pages +
indexStat.empty_pages) * BLCKSZ);
values[j] = palloc(32);
snprintf(values[j++], 32, "%u", indexStat.root_blkno);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT, indexStat.internal_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT, indexStat.leaf_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT, indexStat.empty_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT, indexStat.deleted_pages);
values[j] = palloc(32);
if (indexStat.max_avail > 0)
snprintf(values[j++], 32, "%.2f",
100.0 - (double) indexStat.free_space / (double) indexStat.max_avail * 100.0);
else
snprintf(values[j++], 32, "NaN");
values[j] = palloc(32);
if (indexStat.leaf_pages > 0)
snprintf(values[j++], 32, "%.2f",
(double) indexStat.fragments / (double) indexStat.leaf_pages * 100.0);
else
snprintf(values[j++], 32, "NaN");
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = HeapTupleGetDatum(tuple);
}
PG_RETURN_DATUM(result);
}
/*
* _bt_pagedel() -- Delete a page from the b-tree, if legal to do so.
*
* This action unlinks the page from the b-tree structure, removing all
* pointers leading to it --- but not touching its own left and right links.
* The page cannot be physically reclaimed right away, since other processes
* may currently be trying to follow links leading to the page; they have to
* be allowed to use its right-link to recover. See nbtree/README.
*
* On entry, the target buffer must be pinned and locked (either read or write
* lock is OK). This lock and pin will be dropped before exiting.
*
* The "stack" argument can be a search stack leading (approximately) to the
* target page, or NULL --- outside callers typically pass NULL since they
* have not done such a search, but internal recursion cases pass the stack
* to avoid duplicated search effort.
*
* Returns the number of pages successfully deleted (zero if page cannot
* be deleted now; could be more than one if parent pages were deleted too).
*
* NOTE: this leaks memory. Rather than trying to clean up everything
* carefully, it's better to run it in a temp context that can be reset
* frequently.
*/
int
_bt_pagedel(Relation rel, Buffer buf, BTStack stack, bool vacuum_full)
{
int result;
BlockNumber target,
leftsib,
rightsib,
parent;
OffsetNumber poffset,
maxoff;
uint32 targetlevel,
ilevel;
ItemId itemid;
IndexTuple targetkey,
itup;
ScanKey itup_scankey;
Buffer lbuf,
rbuf,
pbuf;
bool parent_half_dead;
bool parent_one_child;
bool rightsib_empty;
Buffer metabuf = InvalidBuffer;
Page metapg = NULL;
BTMetaPageData *metad = NULL;
Page page;
BTPageOpaque opaque;
MIRROREDLOCK_BUFMGR_MUST_ALREADY_BE_HELD;
// Fetch gp_persistent_relation_node information that will be added to XLOG record.
RelationFetchGpRelationNodeForXLog(rel);
/*
* We can never delete rightmost pages nor root pages. While at it, check
* that page is not already deleted and is empty.
*/
page = BufferGetPage(buf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
if (P_RIGHTMOST(opaque) || P_ISROOT(opaque) || P_ISDELETED(opaque) ||
P_FIRSTDATAKEY(opaque) <= PageGetMaxOffsetNumber(page))
{
/* Should never fail to delete a half-dead page */
Assert(!P_ISHALFDEAD(opaque));
_bt_relbuf(rel, buf);
return 0;
}
/*
* Save info about page, including a copy of its high key (it must have
* one, being non-rightmost).
*/
target = BufferGetBlockNumber(buf);
targetlevel = opaque->btpo.level;
leftsib = opaque->btpo_prev;
itemid = PageGetItemId(page, P_HIKEY);
targetkey = CopyIndexTuple((IndexTuple) PageGetItem(page, itemid));
/*
* To avoid deadlocks, we'd better drop the target page lock before going
* further.
*/
_bt_relbuf(rel, buf);
/*
* We need an approximate pointer to the page's parent page. We use the
* standard search mechanism to search for the page's high key; this will
* give us a link to either the current parent or someplace to its left
* (if there are multiple equal high keys). In recursion cases, the
* caller already generated a search stack and we can just re-use that
* work.
*/
if (stack == NULL)
{
if (!InRecovery)
{
/* we need an insertion scan key to do our search, so build one */
itup_scankey = _bt_mkscankey(rel, targetkey);
/* find the leftmost leaf page containing this key */
stack = _bt_search(rel, rel->rd_rel->relnatts, itup_scankey, false,
&lbuf, BT_READ);
/* don't need a pin on that either */
_bt_relbuf(rel, lbuf);
/*
* If we are trying to delete an interior page, _bt_search did
* more than we needed. Locate the stack item pointing to our
* parent level.
*/
ilevel = 0;
for (;;)
{
if (stack == NULL)
elog(ERROR, "not enough stack items");
if (ilevel == targetlevel)
break;
stack = stack->bts_parent;
ilevel++;
}
}
else
{
/*
* During WAL recovery, we can't use _bt_search (for one reason,
* it might invoke user-defined comparison functions that expect
* facilities not available in recovery mode). Instead, just set
* up a dummy stack pointing to the left end of the parent tree
* level, from which _bt_getstackbuf will walk right to the parent
* page. Painful, but we don't care too much about performance in
* this scenario.
*/
pbuf = _bt_get_endpoint(rel, targetlevel + 1, false);
stack = (BTStack) palloc(sizeof(BTStackData));
stack->bts_blkno = BufferGetBlockNumber(pbuf);
stack->bts_offset = InvalidOffsetNumber;
/* bts_btentry will be initialized below */
stack->bts_parent = NULL;
_bt_relbuf(rel, pbuf);
}
}
/*
* We cannot delete a page that is the rightmost child of its immediate
* parent, unless it is the only child --- in which case the parent has to
* be deleted too, and the same condition applies recursively to it. We
* have to check this condition all the way up before trying to delete. We
* don't need to re-test when deleting a non-leaf page, though.
*/
if (targetlevel == 0 &&
!_bt_parent_deletion_safe(rel, target, stack))
return 0;
/*
* We have to lock the pages we need to modify in the standard order:
* moving right, then up. Else we will deadlock against other writers.
*
* So, we need to find and write-lock the current left sibling of the
* target page. The sibling that was current a moment ago could have
* split, so we may have to move right. This search could fail if either
* the sibling or the target page was deleted by someone else meanwhile;
* if so, give up. (Right now, that should never happen, since page
* deletion is only done in VACUUM and there shouldn't be multiple VACUUMs
* concurrently on the same table.)
*/
if (leftsib != P_NONE)
{
lbuf = _bt_getbuf(rel, leftsib, BT_WRITE);
page = BufferGetPage(lbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
while (P_ISDELETED(opaque) || opaque->btpo_next != target)
{
/* step right one page */
leftsib = opaque->btpo_next;
_bt_relbuf(rel, lbuf);
if (leftsib == P_NONE)
{
elog(LOG, "no left sibling (concurrent deletion?) in \"%s\"",
RelationGetRelationName(rel));
return 0;
}
lbuf = _bt_getbuf(rel, leftsib, BT_WRITE);
page = BufferGetPage(lbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
}
}
else
lbuf = InvalidBuffer;
/*
* Next write-lock the target page itself. It should be okay to take just
* a write lock not a superexclusive lock, since no scans would stop on an
* empty page.
*/
buf = _bt_getbuf(rel, target, BT_WRITE);
page = BufferGetPage(buf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
/*
* Check page is still empty etc, else abandon deletion. The empty check
* is necessary since someone else might have inserted into it while we
* didn't have it locked; the others are just for paranoia's sake.
*/
if (P_RIGHTMOST(opaque) || P_ISROOT(opaque) || P_ISDELETED(opaque) ||
P_FIRSTDATAKEY(opaque) <= PageGetMaxOffsetNumber(page))
{
_bt_relbuf(rel, buf);
if (BufferIsValid(lbuf))
_bt_relbuf(rel, lbuf);
return 0;
}
if (opaque->btpo_prev != leftsib)
elog(ERROR, "left link changed unexpectedly in block %u of index \"%s\"",
target, RelationGetRelationName(rel));
/*
* And next write-lock the (current) right sibling.
*/
rightsib = opaque->btpo_next;
rbuf = _bt_getbuf(rel, rightsib, BT_WRITE);
page = BufferGetPage(rbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
if (opaque->btpo_prev != target)
elog(ERROR, "right sibling's left-link doesn't match: "
"block %u links to %u instead of expected %u in index \"%s\"",
rightsib, opaque->btpo_prev, target,
RelationGetRelationName(rel));
/*
* Next find and write-lock the current parent of the target page. This is
* essentially the same as the corresponding step of splitting.
*/
ItemPointerSet(&(stack->bts_btentry.t_tid), target, P_HIKEY);
pbuf = _bt_getstackbuf(rel, stack, BT_WRITE);
if (pbuf == InvalidBuffer)
elog(ERROR, "failed to re-find parent key in index \"%s\" for deletion target page %u",
RelationGetRelationName(rel), target);
parent = stack->bts_blkno;
poffset = stack->bts_offset;
/*
* If the target is the rightmost child of its parent, then we can't
* delete, unless it's also the only child --- in which case the parent
* changes to half-dead status. The "can't delete" case should have been
* detected by _bt_parent_deletion_safe, so complain if we see it now.
*/
page = BufferGetPage(pbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
maxoff = PageGetMaxOffsetNumber(page);
parent_half_dead = false;
parent_one_child = false;
if (poffset >= maxoff)
{
if (poffset == P_FIRSTDATAKEY(opaque))
parent_half_dead = true;
else
elog(ERROR, "failed to delete rightmost child %u of block %u in index \"%s\"",
target, parent, RelationGetRelationName(rel));
}
else
{
/* Will there be exactly one child left in this parent? */
if (OffsetNumberNext(P_FIRSTDATAKEY(opaque)) == maxoff)
parent_one_child = true;
}
/*
* If we are deleting the next-to-last page on the target's level, then
* the rightsib is a candidate to become the new fast root. (In theory, it
* might be possible to push the fast root even further down, but the odds
* of doing so are slim, and the locking considerations daunting.)
*
* We don't support handling this in the case where the parent is becoming
* half-dead, even though it theoretically could occur.
*
* We can safely acquire a lock on the metapage here --- see comments for
* _bt_newroot().
*/
if (leftsib == P_NONE && !parent_half_dead)
{
page = BufferGetPage(rbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
Assert(opaque->btpo.level == targetlevel);
if (P_RIGHTMOST(opaque))
{
/* rightsib will be the only one left on the level */
metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_WRITE);
metapg = BufferGetPage(metabuf);
metad = BTPageGetMeta(metapg);
/*
* The expected case here is btm_fastlevel == targetlevel+1; if
* the fastlevel is <= targetlevel, something is wrong, and we
* choose to overwrite it to fix it.
*/
if (metad->btm_fastlevel > targetlevel + 1)
{
/* no update wanted */
_bt_relbuf(rel, metabuf);
metabuf = InvalidBuffer;
}
}
}
/*
* Check that the parent-page index items we're about to delete/overwrite
* contain what we expect. This can fail if the index has become
* corrupt for some reason. We want to throw any error before entering
* the critical section --- otherwise it'd be a PANIC.
*
* The test on the target item is just an Assert because _bt_getstackbuf
* should have guaranteed it has the expected contents. The test on the
* next-child downlink is known to sometimes fail in the field, though.
*/
page = BufferGetPage(pbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
#ifdef USE_ASSERT_CHECKING
itemid = PageGetItemId(page, poffset);
itup = (IndexTuple) PageGetItem(page, itemid);
Assert(ItemPointerGetBlockNumber(&(itup->t_tid)) == target);
#endif
if (!parent_half_dead)
{
OffsetNumber nextoffset;
nextoffset = OffsetNumberNext(poffset);
itemid = PageGetItemId(page, nextoffset);
itup = (IndexTuple) PageGetItem(page, itemid);
if (ItemPointerGetBlockNumber(&(itup->t_tid)) != rightsib)
elog(ERROR, "right sibling %u of block %u is not next child %u of block %u in index \"%s\"",
rightsib, target, ItemPointerGetBlockNumber(&(itup->t_tid)),
parent, RelationGetRelationName(rel));
}
/*
* Here we begin doing the deletion.
*/
/* No ereport(ERROR) until changes are logged */
START_CRIT_SECTION();
/*
* Update parent. The normal case is a tad tricky because we want to
* delete the target's downlink and the *following* key. Easiest way is
* to copy the right sibling's downlink over the target downlink, and then
* delete the following item.
*/
if (parent_half_dead)
{
PageIndexTupleDelete(page, poffset);
opaque->btpo_flags |= BTP_HALF_DEAD;
}
else
{
OffsetNumber nextoffset;
itemid = PageGetItemId(page, poffset);
itup = (IndexTuple) PageGetItem(page, itemid);
ItemPointerSet(&(itup->t_tid), rightsib, P_HIKEY);
nextoffset = OffsetNumberNext(poffset);
PageIndexTupleDelete(page, nextoffset);
}
/*
* Update siblings' side-links. Note the target page's side-links will
* continue to point to the siblings. Asserts here are just rechecking
* things we already verified above.
*/
if (BufferIsValid(lbuf))
{
page = BufferGetPage(lbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
Assert(opaque->btpo_next == target);
opaque->btpo_next = rightsib;
}
page = BufferGetPage(rbuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
Assert(opaque->btpo_prev == target);
opaque->btpo_prev = leftsib;
rightsib_empty = (P_FIRSTDATAKEY(opaque) > PageGetMaxOffsetNumber(page));
/*
* Mark the page itself deleted. It can be recycled when all current
* transactions are gone; or immediately if we're doing VACUUM FULL.
*/
page = BufferGetPage(buf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
opaque->btpo_flags &= ~BTP_HALF_DEAD;
opaque->btpo_flags |= BTP_DELETED;
opaque->btpo.xact =
vacuum_full ? FrozenTransactionId : ReadNewTransactionId();
/* And update the metapage, if needed */
if (BufferIsValid(metabuf))
{
metad->btm_fastroot = rightsib;
metad->btm_fastlevel = targetlevel;
MarkBufferDirty(metabuf);
}
/* Must mark buffers dirty before XLogInsert */
MarkBufferDirty(pbuf);
MarkBufferDirty(rbuf);
MarkBufferDirty(buf);
if (BufferIsValid(lbuf))
MarkBufferDirty(lbuf);
/* XLOG stuff */
if (!rel->rd_istemp)
{
xl_btree_delete_page xlrec;
xl_btree_metadata xlmeta;
uint8 xlinfo;
XLogRecPtr recptr;
XLogRecData rdata[5];
XLogRecData *nextrdata;
xl_btreetid_set(&(xlrec.target), rel, parent, poffset);
xlrec.deadblk = target;
xlrec.leftblk = leftsib;
xlrec.rightblk = rightsib;
rdata[0].data = (char *) &xlrec;
rdata[0].len = SizeOfBtreeDeletePage;
rdata[0].buffer = InvalidBuffer;
rdata[0].next = nextrdata = &(rdata[1]);
if (BufferIsValid(metabuf))
{
xlmeta.root = metad->btm_root;
xlmeta.level = metad->btm_level;
xlmeta.fastroot = metad->btm_fastroot;
xlmeta.fastlevel = metad->btm_fastlevel;
nextrdata->data = (char *) &xlmeta;
nextrdata->len = sizeof(xl_btree_metadata);
nextrdata->buffer = InvalidBuffer;
nextrdata->next = nextrdata + 1;
nextrdata++;
xlinfo = XLOG_BTREE_DELETE_PAGE_META;
}
else if (parent_half_dead)
xlinfo = XLOG_BTREE_DELETE_PAGE_HALF;
else
xlinfo = XLOG_BTREE_DELETE_PAGE;
nextrdata->data = NULL;
nextrdata->len = 0;
nextrdata->next = nextrdata + 1;
nextrdata->buffer = pbuf;
nextrdata->buffer_std = true;
nextrdata++;
nextrdata->data = NULL;
nextrdata->len = 0;
nextrdata->buffer = rbuf;
nextrdata->buffer_std = true;
nextrdata->next = NULL;
if (BufferIsValid(lbuf))
{
nextrdata->next = nextrdata + 1;
nextrdata++;
nextrdata->data = NULL;
nextrdata->len = 0;
nextrdata->buffer = lbuf;
nextrdata->buffer_std = true;
nextrdata->next = NULL;
}
recptr = XLogInsert(RM_BTREE_ID, xlinfo, rdata);
if (BufferIsValid(metabuf))
{
PageSetLSN(metapg, recptr);
PageSetTLI(metapg, ThisTimeLineID);
}
page = BufferGetPage(pbuf);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
page = BufferGetPage(rbuf);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
page = BufferGetPage(buf);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
if (BufferIsValid(lbuf))
{
page = BufferGetPage(lbuf);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
}
}
END_CRIT_SECTION();
/* release metapage; send out relcache inval if metapage changed */
if (BufferIsValid(metabuf))
{
CacheInvalidateRelcache(rel);
_bt_relbuf(rel, metabuf);
}
/* can always release leftsib immediately */
if (BufferIsValid(lbuf))
_bt_relbuf(rel, lbuf);
/*
* If parent became half dead, recurse to delete it. Otherwise, if right
* sibling is empty and is now the last child of the parent, recurse to
* try to delete it. (These cases cannot apply at the same time, though
* the second case might itself recurse to the first.)
*
* When recursing to parent, we hold the lock on the target page until
* done. This delays any insertions into the keyspace that was just
* effectively reassigned to the parent's right sibling. If we allowed
* that, and there were enough such insertions before we finish deleting
* the parent, page splits within that keyspace could lead to inserting
* out-of-order keys into the grandparent level. It is thought that that
* wouldn't have any serious consequences, but it still seems like a
* pretty bad idea.
*/
if (parent_half_dead)
{
/* recursive call will release pbuf */
_bt_relbuf(rel, rbuf);
result = _bt_pagedel(rel, pbuf, stack->bts_parent, vacuum_full) + 1;
_bt_relbuf(rel, buf);
}
else if (parent_one_child && rightsib_empty)
{
_bt_relbuf(rel, pbuf);
_bt_relbuf(rel, buf);
/* recursive call will release rbuf */
result = _bt_pagedel(rel, rbuf, stack, vacuum_full) + 1;
}
else
{
_bt_relbuf(rel, pbuf);
_bt_relbuf(rel, buf);
_bt_relbuf(rel, rbuf);
result = 1;
}
return result;
}
/* ------------------------------------------------
* bt_metap()
*
* Get a btree meta-page information
*
* Usage: SELECT * FROM bt_metap('t1_pkey')
* ------------------------------------------------
*/
Datum
bt_metap(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_P(0);
Buffer buffer;
Relation rel;
RangeVar *relrv;
Datum result;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "bt_metap() can be used only on b-tree index.");
buffer = ReadBuffer(rel, 0);
{
BTMetaPageData *metad;
TupleDesc tupleDesc;
int j;
char *values[BTMETAP_NCOLUMNS];
HeapTuple tuple;
Page page = BufferGetPage(buffer);
metad = BTPageGetMeta(page);
tupleDesc = RelationNameGetTupleDesc(BTMETAP_TYPE);
j = 0;
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_magic);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_version);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_root);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_level);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_fastroot);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_fastlevel);
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = TupleGetDatum(TupleDescGetSlot(tupleDesc), tuple);
}
ReleaseBuffer(buffer);
relation_close(rel, AccessShareLock);
PG_RETURN_DATUM(result);
}
/*
* _bt_getroot() -- Get the root page of the btree.
*
* Since the root page can move around the btree file, we have to read
* its location from the metadata page, and then read the root page
* itself. If no root page exists yet, we have to create one. The
* standard class of race conditions exists here; I think I covered
* them all in the Hopi Indian rain dance of lock requests below.
*
* The access type parameter (BT_READ or BT_WRITE) controls whether
* a new root page will be created or not. If access = BT_READ,
* and no root page exists, we just return InvalidBuffer. For
* BT_WRITE, we try to create the root page if it doesn't exist.
* NOTE that the returned root page will have only a read lock set
* on it even if access = BT_WRITE!
*
* The returned page is not necessarily the true root --- it could be
* a "fast root" (a page that is alone in its level due to deletions).
* Also, if the root page is split while we are "in flight" to it,
* what we will return is the old root, which is now just the leftmost
* page on a probably-not-very-wide level. For most purposes this is
* as good as or better than the true root, so we do not bother to
* insist on finding the true root. We do, however, guarantee to
* return a live (not deleted or half-dead) page.
*
* On successful return, the root page is pinned and read-locked.
* The metadata page is not locked or pinned on exit.
*/
Buffer
_bt_getroot(Relation rel, int access)
{
Buffer metabuf;
Page metapg;
BTPageOpaque metaopaque;
Buffer rootbuf;
Page rootpage;
BTPageOpaque rootopaque;
BlockNumber rootblkno;
uint32 rootlevel;
BTMetaPageData *metad;
MIRROREDLOCK_BUFMGR_MUST_ALREADY_BE_HELD;
/*
* Try to use previously-cached metapage data to find the root. This
* normally saves one buffer access per index search, which is a very
* helpful savings in bufmgr traffic and hence contention.
*/
if (rel->rd_amcache != NULL)
{
metad = (BTMetaPageData *) rel->rd_amcache;
/* We shouldn't have cached it if any of these fail */
Assert(metad->btm_magic == BTREE_MAGIC);
Assert(metad->btm_version == BTREE_VERSION);
Assert(metad->btm_root != P_NONE);
rootblkno = metad->btm_fastroot;
Assert(rootblkno != P_NONE);
rootlevel = metad->btm_fastlevel;
rootbuf = _bt_getbuf(rel, rootblkno, BT_READ);
rootpage = BufferGetPage(rootbuf);
rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage);
/*
* Since the cache might be stale, we check the page more carefully
* here than normal. We *must* check that it's not deleted. If it's
* not alone on its level, then we reject too --- this may be overly
* paranoid but better safe than sorry. Note we don't check P_ISROOT,
* because that's not set in a "fast root".
*/
if (!P_IGNORE(rootopaque) &&
rootopaque->btpo.level == rootlevel &&
P_LEFTMOST(rootopaque) &&
P_RIGHTMOST(rootopaque))
{
/* OK, accept cached page as the root */
return rootbuf;
}
_bt_relbuf(rel, rootbuf);
/* Cache is stale, throw it away */
if (rel->rd_amcache)
pfree(rel->rd_amcache);
rel->rd_amcache = NULL;
}
metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_READ);
metapg = BufferGetPage(metabuf);
metaopaque = (BTPageOpaque) PageGetSpecialPointer(metapg);
metad = BTPageGetMeta(metapg);
/* sanity-check the metapage */
if (!(metaopaque->btpo_flags & BTP_META) ||
metad->btm_magic != BTREE_MAGIC)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("index \"%s\" is not a btree",
RelationGetRelationName(rel))));
if (metad->btm_version != BTREE_VERSION)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("version mismatch in index \"%s\": file version %d, code version %d",
RelationGetRelationName(rel),
metad->btm_version, BTREE_VERSION)));
/* if no root page initialized yet, do it */
if (metad->btm_root == P_NONE)
{
/* If access = BT_READ, caller doesn't want us to create root yet */
if (access == BT_READ)
{
_bt_relbuf(rel, metabuf);
return InvalidBuffer;
}
// Fetch gp_persistent_relation_node information that will be added to XLOG record.
RelationFetchGpRelationNodeForXLog(rel);
/* trade in our read lock for a write lock */
LockBuffer(metabuf, BUFFER_LOCK_UNLOCK);
LockBuffer(metabuf, BT_WRITE);
/*
* Race condition: if someone else initialized the metadata between
* the time we released the read lock and acquired the write lock, we
* must avoid doing it again.
*/
if (metad->btm_root != P_NONE)
{
/*
* Metadata initialized by someone else. In order to guarantee no
* deadlocks, we have to release the metadata page and start all
* over again. (Is that really true? But it's hardly worth trying
* to optimize this case.)
*/
_bt_relbuf(rel, metabuf);
return _bt_getroot(rel, access);
}
/*
* Get, initialize, write, and leave a lock of the appropriate type on
* the new root page. Since this is the first page in the tree, it's
* a leaf as well as the root.
*/
rootbuf = _bt_getbuf(rel, P_NEW, BT_WRITE);
rootblkno = BufferGetBlockNumber(rootbuf);
rootpage = BufferGetPage(rootbuf);
rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage);
rootopaque->btpo_prev = rootopaque->btpo_next = P_NONE;
rootopaque->btpo_flags = (BTP_LEAF | BTP_ROOT);
rootopaque->btpo.level = 0;
rootopaque->btpo_cycleid = 0;
/* NO ELOG(ERROR) till meta is updated */
START_CRIT_SECTION();
metad->btm_root = rootblkno;
metad->btm_level = 0;
metad->btm_fastroot = rootblkno;
metad->btm_fastlevel = 0;
MarkBufferDirty(rootbuf);
MarkBufferDirty(metabuf);
/* XLOG stuff */
if (!rel->rd_istemp)
{
xl_btree_newroot xlrec;
XLogRecPtr recptr;
XLogRecData rdata;
xl_btreenode_set(&(xlrec.btreenode), rel);
xlrec.rootblk = rootblkno;
xlrec.level = 0;
rdata.data = (char *) &xlrec;
rdata.len = SizeOfBtreeNewroot;
rdata.buffer = InvalidBuffer;
rdata.next = NULL;
recptr = XLogInsert(RM_BTREE_ID, XLOG_BTREE_NEWROOT, &rdata);
PageSetLSN(rootpage, recptr);
PageSetTLI(rootpage, ThisTimeLineID);
PageSetLSN(metapg, recptr);
PageSetTLI(metapg, ThisTimeLineID);
}
END_CRIT_SECTION();
/*
* Send out relcache inval for metapage change (probably unnecessary
* here, but let's be safe).
*/
CacheInvalidateRelcache(rel);
/*
* swap root write lock for read lock. There is no danger of anyone
* else accessing the new root page while it's unlocked, since no one
* else knows where it is yet.
*/
LockBuffer(rootbuf, BUFFER_LOCK_UNLOCK);
LockBuffer(rootbuf, BT_READ);
/* okay, metadata is correct, release lock on it */
_bt_relbuf(rel, metabuf);
}
else
{
rootblkno = metad->btm_fastroot;
Assert(rootblkno != P_NONE);
rootlevel = metad->btm_fastlevel;
/*
* Cache the metapage data for next time
*/
rel->rd_amcache = MemoryContextAlloc(rel->rd_indexcxt,
sizeof(BTMetaPageData));
memcpy(rel->rd_amcache, metad, sizeof(BTMetaPageData));
/*
* We are done with the metapage; arrange to release it via first
* _bt_relandgetbuf call
*/
rootbuf = metabuf;
for (;;)
{
rootbuf = _bt_relandgetbuf(rel, rootbuf, rootblkno, BT_READ);
rootpage = BufferGetPage(rootbuf);
rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage);
if (!P_IGNORE(rootopaque))
break;
/* it's dead, Jim. step right one page */
if (P_RIGHTMOST(rootopaque))
elog(ERROR, "no live root page found in index \"%s\"",
RelationGetRelationName(rel));
rootblkno = rootopaque->btpo_next;
}
/* Note: can't check btpo.level on deleted pages */
if (rootopaque->btpo.level != rootlevel)
elog(ERROR, "root page %u of index \"%s\" has level %u, expected %u",
rootblkno, RelationGetRelationName(rel),
rootopaque->btpo.level, rootlevel);
}
/*
* By here, we have a pin and read lock on the root page, and no lock set
* on the metadata page. Return the root page's buffer.
*/
return rootbuf;
}
uint32 check_index_page(Relation rel, PageHeader header, char *buffer, int block) {
uint32 nerrs = 0;
BTPageOpaque opaque = NULL;
/* check basic page header */
nerrs += check_page_header(header, block);
/* (block==0) means it's a meta-page, otherwise it's a regular index-page */
if (block == BTREE_METAPAGE) {
BTMetaPageData * mpdata = BTPageGetMeta(buffer);
ereport(DEBUG2, (errmsg("[%d] is a meta-page [magic=%d, version=%d]", block, mpdata->btm_magic, mpdata->btm_version)));
if (mpdata->btm_magic != BTREE_MAGIC) {
ereport(WARNING,(errmsg("[%d] metapage contains invalid magic number %d (should be %d)", block, mpdata->btm_magic, BTREE_MAGIC)));
nerrs++;
}
if (mpdata->btm_version != BTREE_VERSION) {
ereport(WARNING,(errmsg("[%d] metapage contains invalid version %d (should be %d)", block, mpdata->btm_version, BTREE_VERSION)));
nerrs++;
}
/* FIXME Check that the btm_root/btm_fastroot is between 1 and number of index blocks */
/* FIXME Check that the btm_level/btm_fastlevel is equal to the level fo the root block */
} else {
opaque = (BTPageOpaque)(buffer + header->pd_special);
/* check there's enough space for index-relevant data */
if (header->pd_special > BLCKSZ - sizeof(BTPageOpaque)) {
ereport(WARNING,
(errmsg("[%d] there's not enough special space for index data (%d > %d)",
block,
(int) sizeof(BTPageOpaque),
BLCKSZ - header->pd_special)));
nerrs++;
}
/*
* if the page is a leaf page, then level needs to be 0. Otherwise,
* it should be > 0. Deleted pages don't have a level, the level
* field is interleaved with an xid.
*/
if (!P_ISDELETED(opaque))
{
if (P_ISLEAF(opaque))
{
if (opaque-> btpo.level != 0) {
ereport(WARNING,
(errmsg("[%d] is leaf page, but level %d is not zero",
block, opaque->btpo.level)));
nerrs++;
}
}
else
{
if (opaque-> btpo.level == 0) {
ereport(WARNING,
(errmsg("[%d] is a non-leaf page, but level is zero",
block)));
nerrs++;
}
}
}
}
return nerrs;
}