/*
* Visibility map information for a single block of a relation.
*
* Note: the VM code will silently return zeroes for pages past the end
* of the map, so we allow probes up to MaxBlockNumber regardless of the
* actual relation size.
*/
Datum
pg_visibility_map(PG_FUNCTION_ARGS)
{
Oid relid = PG_GETARG_OID(0);
int64 blkno = PG_GETARG_INT64(1);
int32 mapbits;
Relation rel;
Buffer vmbuffer = InvalidBuffer;
TupleDesc tupdesc;
Datum values[2];
bool nulls[2];
rel = relation_open(relid, AccessShareLock);
/* Only some relkinds have a visibility map */
check_relation_relkind(rel);
if (blkno < 0 || blkno > MaxBlockNumber)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("invalid block number")));
tupdesc = pg_visibility_tupdesc(false, false);
MemSet(nulls, 0, sizeof(nulls));
mapbits = (int32) visibilitymap_get_status(rel, blkno, &vmbuffer);
if (vmbuffer != InvalidBuffer)
ReleaseBuffer(vmbuffer);
values[0] = BoolGetDatum((mapbits & VISIBILITYMAP_ALL_VISIBLE) != 0);
values[1] = BoolGetDatum((mapbits & VISIBILITYMAP_ALL_FROZEN) != 0);
relation_close(rel, AccessShareLock);
PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
}
/* Must keep superuser() check, see above. */
Datum
pg_relpagesbyid(PG_FUNCTION_ARGS)
{
Oid relid = PG_GETARG_OID(0);
int64 relpages;
Relation rel;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
rel = relation_open(relid, AccessShareLock);
/* only some relkinds have storage */
check_relation_relkind(rel);
/* note: this will work OK on non-local temp tables */
relpages = RelationGetNumberOfBlocks(rel);
relation_close(rel, AccessShareLock);
PG_RETURN_INT64(relpages);
}
/* --------------------------------------------------------
* pg_relpages()
*
* Get the number of pages of the table/index.
*
* Usage: SELECT pg_relpages('t1');
* SELECT pg_relpages('t1_pkey');
*
* Must keep superuser() check, see above.
* --------------------------------------------------------
*/
Datum
pg_relpages(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_PP(0);
int64 relpages;
Relation rel;
RangeVar *relrv;
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);
/* only some relkinds have storage */
check_relation_relkind(rel);
/* note: this will work OK on non-local temp tables */
relpages = RelationGetNumberOfBlocks(rel);
relation_close(rel, AccessShareLock);
PG_RETURN_INT64(relpages);
}
/*
* Remove the visibility map fork for a relation. If there turn out to be
* any bugs in the visibility map code that require rebuilding the VM, this
* provides users with a way to do it that is cleaner than shutting down the
* server and removing files by hand.
*
* This is a cut-down version of RelationTruncate.
*/
Datum
pg_truncate_visibility_map(PG_FUNCTION_ARGS)
{
Oid relid = PG_GETARG_OID(0);
Relation rel;
rel = relation_open(relid, AccessExclusiveLock);
/* Only some relkinds have a visibility map */
check_relation_relkind(rel);
RelationOpenSmgr(rel);
rel->rd_smgr->smgr_vm_nblocks = InvalidBlockNumber;
visibilitymap_truncate(rel, 0);
if (RelationNeedsWAL(rel))
{
xl_smgr_truncate xlrec;
xlrec.blkno = 0;
xlrec.rnode = rel->rd_node;
xlrec.flags = SMGR_TRUNCATE_VM;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, sizeof(xlrec));
XLogInsert(RM_SMGR_ID, XLOG_SMGR_TRUNCATE | XLR_SPECIAL_REL_UPDATE);
}
/*
* Release the lock right away, not at commit time.
*
* It would be a problem to release the lock prior to commit if this
* truncate operation sends any transactional invalidation messages. Other
* backends would potentially be able to lock the relation without
* processing them in the window of time between when we release the lock
* here and when we sent the messages at our eventual commit. However,
* we're currently only sending a non-transactional smgr invalidation,
* which will have been posted to shared memory immediately from within
* visibilitymap_truncate. Therefore, there should be no race here.
*
* The reason why it's desirable to release the lock early here is because
* of the possibility that someone will need to use this to blow away many
* visibility map forks at once. If we can't release the lock until
* commit time, the transaction doing this will accumulate
* AccessExclusiveLocks on all of those relations at the same time, which
* is undesirable. However, if this turns out to be unsafe we may have no
* choice...
*/
relation_close(rel, AccessExclusiveLock);
/* Nothing to return. */
PG_RETURN_VOID();
}
/*
* Visibility map information for a single block of a relation, plus the
* page-level information for the same block.
*/
Datum
pg_visibility(PG_FUNCTION_ARGS)
{
Oid relid = PG_GETARG_OID(0);
int64 blkno = PG_GETARG_INT64(1);
int32 mapbits;
Relation rel;
Buffer vmbuffer = InvalidBuffer;
Buffer buffer;
Page page;
TupleDesc tupdesc;
Datum values[3];
bool nulls[3];
rel = relation_open(relid, AccessShareLock);
/* Only some relkinds have a visibility map */
check_relation_relkind(rel);
if (blkno < 0 || blkno > MaxBlockNumber)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("invalid block number")));
tupdesc = pg_visibility_tupdesc(false, true);
MemSet(nulls, 0, sizeof(nulls));
mapbits = (int32) visibilitymap_get_status(rel, blkno, &vmbuffer);
if (vmbuffer != InvalidBuffer)
ReleaseBuffer(vmbuffer);
values[0] = BoolGetDatum((mapbits & VISIBILITYMAP_ALL_VISIBLE) != 0);
values[1] = BoolGetDatum((mapbits & VISIBILITYMAP_ALL_FROZEN) != 0);
/* Here we have to explicitly check rel size ... */
if (blkno < RelationGetNumberOfBlocks(rel))
{
buffer = ReadBuffer(rel, blkno);
LockBuffer(buffer, BUFFER_LOCK_SHARE);
page = BufferGetPage(buffer);
values[2] = BoolGetDatum(PageIsAllVisible(page));
UnlockReleaseBuffer(buffer);
}
else
{
/* As with the vismap, silently return 0 for pages past EOF */
values[2] = BoolGetDatum(false);
}
relation_close(rel, AccessShareLock);
PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
}
/*
* Count the number of all-visible and all-frozen pages in the visibility
* map for a particular relation.
*/
Datum
pg_visibility_map_summary(PG_FUNCTION_ARGS)
{
Oid relid = PG_GETARG_OID(0);
Relation rel;
BlockNumber nblocks;
BlockNumber blkno;
Buffer vmbuffer = InvalidBuffer;
int64 all_visible = 0;
int64 all_frozen = 0;
TupleDesc tupdesc;
Datum values[2];
bool nulls[2];
rel = relation_open(relid, AccessShareLock);
/* Only some relkinds have a visibility map */
check_relation_relkind(rel);
nblocks = RelationGetNumberOfBlocks(rel);
for (blkno = 0; blkno < nblocks; ++blkno)
{
int32 mapbits;
/* Make sure we are interruptible. */
CHECK_FOR_INTERRUPTS();
/* Get map info. */
mapbits = (int32) visibilitymap_get_status(rel, blkno, &vmbuffer);
if ((mapbits & VISIBILITYMAP_ALL_VISIBLE) != 0)
++all_visible;
if ((mapbits & VISIBILITYMAP_ALL_FROZEN) != 0)
++all_frozen;
}
/* Clean up. */
if (vmbuffer != InvalidBuffer)
ReleaseBuffer(vmbuffer);
relation_close(rel, AccessShareLock);
tupdesc = CreateTemplateTupleDesc(2, false);
TupleDescInitEntry(tupdesc, (AttrNumber) 1, "all_visible", INT8OID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 2, "all_frozen", INT8OID, -1, 0);
tupdesc = BlessTupleDesc(tupdesc);
MemSet(nulls, 0, sizeof(nulls));
values[0] = Int64GetDatum(all_visible);
values[1] = Int64GetDatum(all_frozen);
PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
}
/* No need for superuser checks in v1.5, see above */
Datum
pg_relpagesbyid_v1_5(PG_FUNCTION_ARGS)
{
Oid relid = PG_GETARG_OID(0);
int64 relpages;
Relation rel;
rel = relation_open(relid, AccessShareLock);
/* only some relkinds have storage */
check_relation_relkind(rel);
/* note: this will work OK on non-local temp tables */
relpages = RelationGetNumberOfBlocks(rel);
relation_close(rel, AccessShareLock);
PG_RETURN_INT64(relpages);
}
/* No need for superuser checks in v1.5, see above */
Datum
pg_relpages_v1_5(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_PP(0);
int64 relpages;
Relation rel;
RangeVar *relrv;
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
/* only some relkinds have storage */
check_relation_relkind(rel);
/* note: this will work OK on non-local temp tables */
relpages = RelationGetNumberOfBlocks(rel);
relation_close(rel, AccessShareLock);
PG_RETURN_INT64(relpages);
}
/*
* Returns a list of items whose visibility map information does not match
* the status of the tuples on the page.
*
* If all_visible is passed as true, this will include all items which are
* on pages marked as all-visible in the visibility map but which do not
* seem to in fact be all-visible.
*
* If all_frozen is passed as true, this will include all items which are
* on pages marked as all-frozen but which do not seem to in fact be frozen.
*
* Checks relkind of relid and will throw an error if the relation does not
* have a VM.
*/
static corrupt_items *
collect_corrupt_items(Oid relid, bool all_visible, bool all_frozen)
{
Relation rel;
BlockNumber nblocks;
corrupt_items *items;
BlockNumber blkno;
Buffer vmbuffer = InvalidBuffer;
BufferAccessStrategy bstrategy = GetAccessStrategy(BAS_BULKREAD);
TransactionId OldestXmin = InvalidTransactionId;
if (all_visible)
{
/* Don't pass rel; that will fail in recovery. */
OldestXmin = GetOldestXmin(NULL, PROCARRAY_FLAGS_VACUUM);
}
rel = relation_open(relid, AccessShareLock);
/* Only some relkinds have a visibility map */
check_relation_relkind(rel);
nblocks = RelationGetNumberOfBlocks(rel);
/*
* Guess an initial array size. We don't expect many corrupted tuples, so
* start with a small array. This function uses the "next" field to track
* the next offset where we can store an item (which is the same thing as
* the number of items found so far) and the "count" field to track the
* number of entries allocated. We'll repurpose these fields before
* returning.
*/
items = palloc0(sizeof(corrupt_items));
items->next = 0;
items->count = 64;
items->tids = palloc(items->count * sizeof(ItemPointerData));
/* Loop over every block in the relation. */
for (blkno = 0; blkno < nblocks; ++blkno)
{
bool check_frozen = false;
bool check_visible = false;
Buffer buffer;
Page page;
OffsetNumber offnum,
maxoff;
/* Make sure we are interruptible. */
CHECK_FOR_INTERRUPTS();
/* Use the visibility map to decide whether to check this page. */
if (all_frozen && VM_ALL_FROZEN(rel, blkno, &vmbuffer))
check_frozen = true;
if (all_visible && VM_ALL_VISIBLE(rel, blkno, &vmbuffer))
check_visible = true;
if (!check_visible && !check_frozen)
continue;
/* Read and lock the page. */
buffer = ReadBufferExtended(rel, MAIN_FORKNUM, blkno, RBM_NORMAL,
bstrategy);
LockBuffer(buffer, BUFFER_LOCK_SHARE);
page = BufferGetPage(buffer);
maxoff = PageGetMaxOffsetNumber(page);
/*
* The visibility map bits might have changed while we were acquiring
* the page lock. Recheck to avoid returning spurious results.
*/
if (check_frozen && !VM_ALL_FROZEN(rel, blkno, &vmbuffer))
check_frozen = false;
if (check_visible && !VM_ALL_VISIBLE(rel, blkno, &vmbuffer))
check_visible = false;
if (!check_visible && !check_frozen)
{
UnlockReleaseBuffer(buffer);
continue;
}
/* Iterate over each tuple on the page. */
for (offnum = FirstOffsetNumber;
offnum <= maxoff;
offnum = OffsetNumberNext(offnum))
{
HeapTupleData tuple;
ItemId itemid;
itemid = PageGetItemId(page, offnum);
/* Unused or redirect line pointers are of no interest. */
if (!ItemIdIsUsed(itemid) || ItemIdIsRedirected(itemid))
continue;
/* Dead line pointers are neither all-visible nor frozen. */
if (ItemIdIsDead(itemid))
{
ItemPointerSet(&(tuple.t_self), blkno, offnum);
record_corrupt_item(items, &tuple.t_self);
continue;
}
/* Initialize a HeapTupleData structure for checks below. */
ItemPointerSet(&(tuple.t_self), blkno, offnum);
tuple.t_data = (HeapTupleHeader) PageGetItem(page, itemid);
tuple.t_len = ItemIdGetLength(itemid);
tuple.t_tableOid = relid;
/*
* If we're checking whether the page is all-visible, we expect
* the tuple to be all-visible.
*/
if (check_visible &&
!tuple_all_visible(&tuple, OldestXmin, buffer))
{
TransactionId RecomputedOldestXmin;
/*
* Time has passed since we computed OldestXmin, so it's
* possible that this tuple is all-visible in reality even
* though it doesn't appear so based on our
* previously-computed value. Let's compute a new value so we
* can be certain whether there is a problem.
*
* From a concurrency point of view, it sort of sucks to
* retake ProcArrayLock here while we're holding the buffer
* exclusively locked, but it should be safe against
* deadlocks, because surely GetOldestXmin() should never take
* a buffer lock. And this shouldn't happen often, so it's
* worth being careful so as to avoid false positives.
*/
RecomputedOldestXmin = GetOldestXmin(NULL, PROCARRAY_FLAGS_VACUUM);
if (!TransactionIdPrecedes(OldestXmin, RecomputedOldestXmin))
record_corrupt_item(items, &tuple.t_self);
else
{
OldestXmin = RecomputedOldestXmin;
if (!tuple_all_visible(&tuple, OldestXmin, buffer))
record_corrupt_item(items, &tuple.t_self);
}
}
/*
* If we're checking whether the page is all-frozen, we expect the
* tuple to be in a state where it will never need freezing.
*/
if (check_frozen)
{
if (heap_tuple_needs_eventual_freeze(tuple.t_data))
record_corrupt_item(items, &tuple.t_self);
}
}
UnlockReleaseBuffer(buffer);
}
/* Clean up. */
if (vmbuffer != InvalidBuffer)
ReleaseBuffer(vmbuffer);
relation_close(rel, AccessShareLock);
/*
* Before returning, repurpose the fields to match caller's expectations.
* next is now the next item that should be read (rather than written) and
* count is now the number of items we wrote (rather than the number we
* allocated).
*/
items->count = items->next;
items->next = 0;
return items;
}
/*
* Collect visibility data about a relation.
*
* Checks relkind of relid and will throw an error if the relation does not
* have a VM.
*/
static vbits *
collect_visibility_data(Oid relid, bool include_pd)
{
Relation rel;
BlockNumber nblocks;
vbits *info;
BlockNumber blkno;
Buffer vmbuffer = InvalidBuffer;
BufferAccessStrategy bstrategy = GetAccessStrategy(BAS_BULKREAD);
rel = relation_open(relid, AccessShareLock);
/* Only some relkinds have a visibility map */
check_relation_relkind(rel);
nblocks = RelationGetNumberOfBlocks(rel);
info = palloc0(offsetof(vbits, bits) + nblocks);
info->next = 0;
info->count = nblocks;
for (blkno = 0; blkno < nblocks; ++blkno)
{
int32 mapbits;
/* Make sure we are interruptible. */
CHECK_FOR_INTERRUPTS();
/* Get map info. */
mapbits = (int32) visibilitymap_get_status(rel, blkno, &vmbuffer);
if ((mapbits & VISIBILITYMAP_ALL_VISIBLE) != 0)
info->bits[blkno] |= (1 << 0);
if ((mapbits & VISIBILITYMAP_ALL_FROZEN) != 0)
info->bits[blkno] |= (1 << 1);
/*
* Page-level data requires reading every block, so only get it if the
* caller needs it. Use a buffer access strategy, too, to prevent
* cache-trashing.
*/
if (include_pd)
{
Buffer buffer;
Page page;
buffer = ReadBufferExtended(rel, MAIN_FORKNUM, blkno, RBM_NORMAL,
bstrategy);
LockBuffer(buffer, BUFFER_LOCK_SHARE);
page = BufferGetPage(buffer);
if (PageIsAllVisible(page))
info->bits[blkno] |= (1 << 2);
UnlockReleaseBuffer(buffer);
}
}
/* Clean up. */
if (vmbuffer != InvalidBuffer)
ReleaseBuffer(vmbuffer);
relation_close(rel, AccessShareLock);
return info;
}