/*
* Collect visibility data about a relation.
*/
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);
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, NULL, NULL, BGP_NO_SNAPSHOT_TEST);
if (PageIsAllVisible(page))
info->bits[blkno] |= (1 << 2);
UnlockReleaseBuffer(buffer);
}
}
/* Clean up. */
if (vmbuffer != InvalidBuffer)
ReleaseBuffer(vmbuffer);
relation_close(rel, AccessShareLock);
return info;
}
/*
* pgstat_index -- returns live/dead tuples info in a generic index
*/
static Datum
pgstat_index(Relation rel, BlockNumber start, pgstat_page pagefn,
FunctionCallInfo fcinfo)
{
BlockNumber nblocks;
BlockNumber blkno;
BufferAccessStrategy bstrategy;
pgstattuple_type stat = {0};
/* prepare access strategy for this index */
bstrategy = GetAccessStrategy(BAS_BULKREAD);
blkno = start;
for (;;)
{
/* Get the current relation length */
LockRelationForExtension(rel, ExclusiveLock);
nblocks = RelationGetNumberOfBlocks(rel);
UnlockRelationForExtension(rel, ExclusiveLock);
/* Quit if we've scanned the whole relation */
if (blkno >= nblocks)
{
stat.table_len = (uint64) nblocks *BLCKSZ;
break;
}
for (; blkno < nblocks; blkno++)
{
CHECK_FOR_INTERRUPTS();
pagefn(&stat, rel, blkno, bstrategy);
}
}
relation_close(rel, AccessShareLock);
return build_pgstattuple_type(&stat, fcinfo);
}
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;
}
/* ------------------------------------------------------
* pgstathashindex()
*
* Usage: SELECT * FROM pgstathashindex('hashindex');
* ------------------------------------------------------
*/
Datum
pgstathashindex(PG_FUNCTION_ARGS)
{
Oid relid = PG_GETARG_OID(0);
BlockNumber nblocks;
BlockNumber blkno;
Relation rel;
HashIndexStat stats;
BufferAccessStrategy bstrategy;
HeapTuple tuple;
TupleDesc tupleDesc;
Datum values[8];
bool nulls[8];
Buffer metabuf;
HashMetaPage metap;
float8 free_percent;
uint64 total_space;
rel = index_open(relid, AccessShareLock);
/* index_open() checks that it's an index */
if (!IS_HASH(rel))
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("relation \"%s\" is not a HASH 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 indexes of other sessions")));
/* Get the information we need from the metapage. */
memset(&stats, 0, sizeof(stats));
metabuf = _hash_getbuf(rel, HASH_METAPAGE, HASH_READ, LH_META_PAGE);
metap = HashPageGetMeta(BufferGetPage(metabuf));
stats.version = metap->hashm_version;
stats.space_per_page = metap->hashm_bsize;
_hash_relbuf(rel, metabuf);
/* Get the current relation length */
nblocks = RelationGetNumberOfBlocks(rel);
/* prepare access strategy for this index */
bstrategy = GetAccessStrategy(BAS_BULKREAD);
/* Start from blkno 1 as 0th block is metapage */
for (blkno = 1; blkno < nblocks; blkno++)
{
Buffer buf;
Page page;
CHECK_FOR_INTERRUPTS();
buf = ReadBufferExtended(rel, MAIN_FORKNUM, blkno, RBM_NORMAL,
bstrategy);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = (Page) BufferGetPage(buf);
if (PageIsNew(page))
stats.unused_pages++;
else if (PageGetSpecialSize(page) !=
MAXALIGN(sizeof(HashPageOpaqueData)))
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("index \"%s\" contains corrupted page at block %u",
RelationGetRelationName(rel),
BufferGetBlockNumber(buf))));
else
{
HashPageOpaque opaque;
int pagetype;
opaque = (HashPageOpaque) PageGetSpecialPointer(page);
pagetype = opaque->hasho_flag & LH_PAGE_TYPE;
if (pagetype == LH_BUCKET_PAGE)
{
stats.bucket_pages++;
GetHashPageStats(page, &stats);
}
else if (pagetype == LH_OVERFLOW_PAGE)
{
stats.overflow_pages++;
GetHashPageStats(page, &stats);
}
else if (pagetype == LH_BITMAP_PAGE)
stats.bitmap_pages++;
else if (pagetype == LH_UNUSED_PAGE)
stats.unused_pages++;
else
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("unexpected page type 0x%04X in HASH index \"%s\" block %u",
opaque->hasho_flag, RelationGetRelationName(rel),
BufferGetBlockNumber(buf))));
}
UnlockReleaseBuffer(buf);
}
/* Done accessing the index */
index_close(rel, AccessShareLock);
/* Count unused pages as free space. */
stats.free_space += stats.unused_pages * stats.space_per_page;
/*
* Total space available for tuples excludes the metapage and the bitmap
* pages.
*/
total_space = (nblocks - (stats.bitmap_pages + 1)) * stats.space_per_page;
if (total_space == 0)
free_percent = 0.0;
else
free_percent = 100.0 * stats.free_space / total_space;
/*
* 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");
tupleDesc = BlessTupleDesc(tupleDesc);
/*
* Build and return the tuple
*/
MemSet(nulls, 0, sizeof(nulls));
values[0] = Int32GetDatum(stats.version);
values[1] = Int64GetDatum((int64) stats.bucket_pages);
values[2] = Int64GetDatum((int64) stats.overflow_pages);
values[3] = Int64GetDatum((int64) stats.bitmap_pages);
values[4] = Int64GetDatum((int64) stats.unused_pages);
values[5] = Int64GetDatum(stats.live_items);
values[6] = Int64GetDatum(stats.dead_items);
values[7] = Float8GetDatum(free_percent);
tuple = heap_form_tuple(tupleDesc, values, nulls);
PG_RETURN_DATUM(HeapTupleGetDatum(tuple));
}
/*
* This function takes an already open relation and scans its pages,
* skipping those that have the corresponding visibility map bit set.
* For pages we skip, we find the free space from the free space map
* and approximate tuple_len on that basis. For the others, we count
* the exact number of dead tuples etc.
*
* This scan is loosely based on vacuumlazy.c:lazy_scan_heap(), but
* we do not try to avoid skipping single pages.
*/
static void
statapprox_heap(Relation rel, output_type *stat)
{
BlockNumber scanned,
nblocks,
blkno;
Buffer vmbuffer = InvalidBuffer;
BufferAccessStrategy bstrategy;
TransactionId OldestXmin;
uint64 misc_count = 0;
OldestXmin = GetOldestXmin(rel, PROCARRAY_FLAGS_VACUUM);
bstrategy = GetAccessStrategy(BAS_BULKREAD);
nblocks = RelationGetNumberOfBlocks(rel);
scanned = 0;
for (blkno = 0; blkno < nblocks; blkno++)
{
Buffer buf;
Page page;
OffsetNumber offnum,
maxoff;
Size freespace;
CHECK_FOR_INTERRUPTS();
/*
* If the page has only visible tuples, then we can find out the free
* space from the FSM and move on.
*/
if (VM_ALL_VISIBLE(rel, blkno, &vmbuffer))
{
freespace = GetRecordedFreeSpace(rel, blkno);
stat->tuple_len += BLCKSZ - freespace;
stat->free_space += freespace;
continue;
}
buf = ReadBufferExtended(rel, MAIN_FORKNUM, blkno,
RBM_NORMAL, bstrategy);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
/*
* It's not safe to call PageGetHeapFreeSpace() on new pages, so we
* treat them as being free space for our purposes.
*/
if (!PageIsNew(page))
stat->free_space += PageGetHeapFreeSpace(page);
else
stat->free_space += BLCKSZ - SizeOfPageHeaderData;
if (PageIsNew(page) || PageIsEmpty(page))
{
UnlockReleaseBuffer(buf);
continue;
}
scanned++;
/*
* Look at each tuple on the page and decide whether it's live or
* dead, then count it and its size. Unlike lazy_scan_heap, we can
* afford to ignore problems and special cases.
*/
maxoff = PageGetMaxOffsetNumber(page);
for (offnum = FirstOffsetNumber;
offnum <= maxoff;
offnum = OffsetNumberNext(offnum))
{
ItemId itemid;
HeapTupleData tuple;
itemid = PageGetItemId(page, offnum);
if (!ItemIdIsUsed(itemid) || ItemIdIsRedirected(itemid) ||
ItemIdIsDead(itemid))
{
continue;
}
Assert(ItemIdIsNormal(itemid));
ItemPointerSet(&(tuple.t_self), blkno, offnum);
tuple.t_data = (HeapTupleHeader) PageGetItem(page, itemid);
tuple.t_len = ItemIdGetLength(itemid);
tuple.t_tableOid = RelationGetRelid(rel);
/*
* We count live and dead tuples, but we also need to add up
* others in order to feed vac_estimate_reltuples.
*/
switch (HeapTupleSatisfiesVacuum(&tuple, OldestXmin, buf))
{
case HEAPTUPLE_RECENTLY_DEAD:
misc_count++;
/* Fall through */
case HEAPTUPLE_DEAD:
stat->dead_tuple_len += tuple.t_len;
stat->dead_tuple_count++;
break;
case HEAPTUPLE_LIVE:
stat->tuple_len += tuple.t_len;
stat->tuple_count++;
break;
case HEAPTUPLE_INSERT_IN_PROGRESS:
case HEAPTUPLE_DELETE_IN_PROGRESS:
misc_count++;
break;
default:
elog(ERROR, "unexpected HeapTupleSatisfiesVacuum result");
break;
}
}
UnlockReleaseBuffer(buf);
}
stat->table_len = (uint64) nblocks *BLCKSZ;
stat->tuple_count = vac_estimate_reltuples(rel, false, nblocks, scanned,
stat->tuple_count + misc_count);
/*
* Calculate percentages if the relation has one or more pages.
*/
if (nblocks != 0)
{
stat->scanned_percent = 100 * scanned / nblocks;
stat->tuple_percent = 100.0 * stat->tuple_len / stat->table_len;
stat->dead_tuple_percent = 100.0 * stat->dead_tuple_len / stat->table_len;
stat->free_percent = 100.0 * stat->free_space / stat->table_len;
}
if (BufferIsValid(vmbuffer))
{
ReleaseBuffer(vmbuffer);
vmbuffer = InvalidBuffer;
}
}
/*
* Insert all matching tuples into to a bitmap.
*/
int64
blgetbitmap(IndexScanDesc scan, TIDBitmap *tbm)
{
int64 ntids = 0;
BlockNumber blkno = BLOOM_HEAD_BLKNO,
npages;
int i;
BufferAccessStrategy bas;
BloomScanOpaque so = (BloomScanOpaque) scan->opaque;
if (so->sign == NULL)
{
/* New search: have to calculate search signature */
ScanKey skey = scan->keyData;
so->sign = palloc0(sizeof(SignType) * so->state.opts.bloomLength);
for (i = 0; i < scan->numberOfKeys; i++)
{
/*
* Assume bloom-indexable operators to be strict, so nothing could
* be found for NULL key.
*/
if (skey->sk_flags & SK_ISNULL)
{
pfree(so->sign);
so->sign = NULL;
return 0;
}
/* Add next value to the signature */
signValue(&so->state, so->sign, skey->sk_argument,
skey->sk_attno - 1);
skey++;
}
}
/*
* We're going to read the whole index. This is why we use appropriate
* buffer access strategy.
*/
bas = GetAccessStrategy(BAS_BULKREAD);
npages = RelationGetNumberOfBlocks(scan->indexRelation);
for (blkno = BLOOM_HEAD_BLKNO; blkno < npages; blkno++)
{
Buffer buffer;
Page page;
buffer = ReadBufferExtended(scan->indexRelation, MAIN_FORKNUM,
blkno, RBM_NORMAL, bas);
LockBuffer(buffer, BUFFER_LOCK_SHARE);
page = BufferGetPage(buffer);
TestForOldSnapshot(scan->xs_snapshot, scan->indexRelation, page);
if (!BloomPageIsDeleted(page))
{
OffsetNumber offset,
maxOffset = BloomPageGetMaxOffset(page);
for (offset = 1; offset <= maxOffset; offset++)
{
BloomTuple *itup = BloomPageGetTuple(&so->state, page, offset);
bool res = true;
/* Check index signature with scan signature */
for (i = 0; i < so->state.opts.bloomLength; i++)
{
if ((itup->sign[i] & so->sign[i]) != so->sign[i])
{
res = false;
break;
}
}
/* Add matching tuples to bitmap */
if (res)
{
tbm_add_tuples(tbm, &itup->heapPtr, 1, true);
ntids++;
}
}
}
UnlockReleaseBuffer(buffer);
CHECK_FOR_INTERRUPTS();
}
FreeAccessStrategy(bas);
return ntids;
}
/*
* 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.
*/
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, true);
}
rel = relation_open(relid, AccessShareLock);
if (rel->rd_rel->relkind != RELKIND_RELATION &&
rel->rd_rel->relkind != RELKIND_MATVIEW &&
rel->rd_rel->relkind != RELKIND_TOASTVALUE)
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("\"%s\" is not a table, materialized view, or TOAST table",
RelationGetRelationName(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, true);
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;
}
/*
* Primary entry point for VACUUM and ANALYZE commands.
*
* relid is normally InvalidOid; if it is not, then it provides the relation
* OID to be processed, and vacstmt->relation is ignored. (The non-invalid
* case is currently only used by autovacuum.)
*
* do_toast is passed as FALSE by autovacuum, because it processes TOAST
* tables separately.
*
* for_wraparound is used by autovacuum to let us know when it's forcing
* a vacuum for wraparound, which should not be auto-cancelled.
*
* bstrategy is normally given as NULL, but in autovacuum it can be passed
* in to use the same buffer strategy object across multiple vacuum() calls.
*
* isTopLevel should be passed down from ProcessUtility.
*
* It is the caller's responsibility that vacstmt and bstrategy
* (if given) be allocated in a memory context that won't disappear
* at transaction commit.
*/
void
vacuum(VacuumStmt *vacstmt, Oid relid, bool do_toast,
BufferAccessStrategy bstrategy, bool for_wraparound, bool isTopLevel)
{
const char *stmttype;
volatile bool all_rels,
in_outer_xact,
use_own_xacts;
List *relations;
/* sanity checks on options */
Assert(vacstmt->options & (VACOPT_VACUUM | VACOPT_ANALYZE));
Assert((vacstmt->options & VACOPT_VACUUM) ||
!(vacstmt->options & (VACOPT_FULL | VACOPT_FREEZE)));
Assert((vacstmt->options & VACOPT_ANALYZE) || vacstmt->va_cols == NIL);
stmttype = (vacstmt->options & VACOPT_VACUUM) ? "VACUUM" : "ANALYZE";
/*
* We cannot run VACUUM inside a user transaction block; if we were inside
* a transaction, then our commit- and start-transaction-command calls
* would not have the intended effect! There are numerous other subtle
* dependencies on this, too.
*
* ANALYZE (without VACUUM) can run either way.
*/
if (vacstmt->options & VACOPT_VACUUM)
{
PreventTransactionChain(isTopLevel, stmttype);
in_outer_xact = false;
}
else
in_outer_xact = IsInTransactionChain(isTopLevel);
/*
* Send info about dead objects to the statistics collector, unless we are
* in autovacuum --- autovacuum.c does this for itself.
*/
if ((vacstmt->options & VACOPT_VACUUM) && !IsAutoVacuumWorkerProcess())
pgstat_vacuum_stat();
/*
* Create special memory context for cross-transaction storage.
*
* Since it is a child of PortalContext, it will go away eventually even
* if we suffer an error; there's no need for special abort cleanup logic.
*/
vac_context = AllocSetContextCreate(PortalContext,
"Vacuum",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
/*
* If caller didn't give us a buffer strategy object, make one in the
* cross-transaction memory context.
*/
if (bstrategy == NULL)
{
MemoryContext old_context = MemoryContextSwitchTo(vac_context);
bstrategy = GetAccessStrategy(BAS_VACUUM);
MemoryContextSwitchTo(old_context);
}
vac_strategy = bstrategy;
/* Remember whether we are processing everything in the DB */
all_rels = (!OidIsValid(relid) && vacstmt->relation == NULL);
/*
* Build list of relations to process, unless caller gave us one. (If we
* build one, we put it in vac_context for safekeeping.)
*/
relations = get_rel_oids(relid, vacstmt->relation);
/*
* Decide whether we need to start/commit our own transactions.
*
* For VACUUM (with or without ANALYZE): always do so, so that we can
* release locks as soon as possible. (We could possibly use the outer
* transaction for a one-table VACUUM, but handling TOAST tables would be
* problematic.)
*
* For ANALYZE (no VACUUM): if inside a transaction block, we cannot
* start/commit our own transactions. Also, there's no need to do so if
* only processing one relation. For multiple relations when not within a
* transaction block, and also in an autovacuum worker, use own
* transactions so we can release locks sooner.
*/
if (vacstmt->options & VACOPT_VACUUM)
use_own_xacts = true;
else
{
Assert(vacstmt->options & VACOPT_ANALYZE);
if (IsAutoVacuumWorkerProcess())
use_own_xacts = true;
else if (in_outer_xact)
use_own_xacts = false;
else if (list_length(relations) > 1)
use_own_xacts = true;
else
use_own_xacts = false;
}
/*
* vacuum_rel expects to be entered with no transaction active; it will
* start and commit its own transaction. But we are called by an SQL
* command, and so we are executing inside a transaction already. We
* commit the transaction started in PostgresMain() here, and start
* another one before exiting to match the commit waiting for us back in
* PostgresMain().
*/
if (use_own_xacts)
{
/* ActiveSnapshot is not set by autovacuum */
if (ActiveSnapshotSet())
PopActiveSnapshot();
/* matches the StartTransaction in PostgresMain() */
CommitTransactionCommand();
}
/* Turn vacuum cost accounting on or off */
PG_TRY();
{
ListCell *cur;
VacuumCostActive = (VacuumCostDelay > 0);
VacuumCostBalance = 0;
/*
* Loop to process each selected relation.
*/
foreach(cur, relations)
{
Oid relid = lfirst_oid(cur);
bool scanned_all = false;
if (vacstmt->options & VACOPT_VACUUM)
vacuum_rel(relid, vacstmt, do_toast, for_wraparound,
&scanned_all);
if (vacstmt->options & VACOPT_ANALYZE)
{
/*
* If using separate xacts, start one for analyze. Otherwise,
* we can use the outer transaction.
*/
if (use_own_xacts)
{
StartTransactionCommand();
/* functions in indexes may want a snapshot set */
PushActiveSnapshot(GetTransactionSnapshot());
}
analyze_rel(relid, vacstmt, vac_strategy, !scanned_all);
if (use_own_xacts)
{
PopActiveSnapshot();
CommitTransactionCommand();
}
}
}
}
/*
* Primary entry point for VACUUM and ANALYZE commands.
*
* options is a bitmask of VacuumOption flags, indicating what to do.
*
* relid, if not InvalidOid, indicate the relation to process; otherwise,
* the RangeVar is used. (The latter must always be passed, because it's
* used for error messages.)
*
* params contains a set of parameters that can be used to customize the
* behavior.
*
* va_cols is a list of columns to analyze, or NIL to process them all.
*
* bstrategy is normally given as NULL, but in autovacuum it can be passed
* in to use the same buffer strategy object across multiple vacuum() calls.
*
* isTopLevel should be passed down from ProcessUtility.
*
* It is the caller's responsibility that all parameters are allocated in a
* memory context that will not disappear at transaction commit.
*/
void
vacuum(int options, RangeVar *relation, Oid relid, VacuumParams *params,
List *va_cols, BufferAccessStrategy bstrategy, bool isTopLevel)
{
const char *stmttype;
volatile bool in_outer_xact,
use_own_xacts;
List *relations;
static bool in_vacuum = false;
Assert(params != NULL);
stmttype = (options & VACOPT_VACUUM) ? "VACUUM" : "ANALYZE";
/*
* We cannot run VACUUM inside a user transaction block; if we were inside
* a transaction, then our commit- and start-transaction-command calls
* would not have the intended effect! There are numerous other subtle
* dependencies on this, too.
*
* ANALYZE (without VACUUM) can run either way.
*/
if (options & VACOPT_VACUUM)
{
PreventTransactionChain(isTopLevel, stmttype);
in_outer_xact = false;
}
else
in_outer_xact = IsInTransactionChain(isTopLevel);
/*
* Due to static variables vac_context, anl_context and vac_strategy,
* vacuum() is not reentrant. This matters when VACUUM FULL or ANALYZE
* calls a hostile index expression that itself calls ANALYZE.
*/
if (in_vacuum)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("%s cannot be executed from VACUUM or ANALYZE",
stmttype)));
/*
* Sanity check DISABLE_PAGE_SKIPPING option.
*/
if ((options & VACOPT_FULL) != 0 &&
(options & VACOPT_DISABLE_PAGE_SKIPPING) != 0)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("VACUUM option DISABLE_PAGE_SKIPPING cannot be used with FULL")));
/*
* Send info about dead objects to the statistics collector, unless we are
* in autovacuum --- autovacuum.c does this for itself.
*/
if ((options & VACOPT_VACUUM) && !IsAutoVacuumWorkerProcess())
pgstat_vacuum_stat();
/*
* Create special memory context for cross-transaction storage.
*
* Since it is a child of PortalContext, it will go away eventually even
* if we suffer an error; there's no need for special abort cleanup logic.
*/
vac_context = AllocSetContextCreate(PortalContext,
"Vacuum",
ALLOCSET_DEFAULT_SIZES);
/*
* If caller didn't give us a buffer strategy object, make one in the
* cross-transaction memory context.
*/
if (bstrategy == NULL)
{
MemoryContext old_context = MemoryContextSwitchTo(vac_context);
bstrategy = GetAccessStrategy(BAS_VACUUM);
MemoryContextSwitchTo(old_context);
}
vac_strategy = bstrategy;
/*
* Build list of relations to process, unless caller gave us one. (If we
* build one, we put it in vac_context for safekeeping.)
*/
relations = get_rel_oids(relid, relation);
/*
* Decide whether we need to start/commit our own transactions.
*
* For VACUUM (with or without ANALYZE): always do so, so that we can
* release locks as soon as possible. (We could possibly use the outer
* transaction for a one-table VACUUM, but handling TOAST tables would be
* problematic.)
*
* For ANALYZE (no VACUUM): if inside a transaction block, we cannot
* start/commit our own transactions. Also, there's no need to do so if
* only processing one relation. For multiple relations when not within a
* transaction block, and also in an autovacuum worker, use own
* transactions so we can release locks sooner.
*/
if (options & VACOPT_VACUUM)
use_own_xacts = true;
else
{
Assert(options & VACOPT_ANALYZE);
if (IsAutoVacuumWorkerProcess())
use_own_xacts = true;
else if (in_outer_xact)
use_own_xacts = false;
else if (list_length(relations) > 1)
use_own_xacts = true;
else
use_own_xacts = false;
}
/*
* vacuum_rel expects to be entered with no transaction active; it will
* start and commit its own transaction. But we are called by an SQL
* command, and so we are executing inside a transaction already. We
* commit the transaction started in PostgresMain() here, and start
* another one before exiting to match the commit waiting for us back in
* PostgresMain().
*/
if (use_own_xacts)
{
Assert(!in_outer_xact);
/* ActiveSnapshot is not set by autovacuum */
if (ActiveSnapshotSet())
PopActiveSnapshot();
/* matches the StartTransaction in PostgresMain() */
CommitTransactionCommand();
}
/* Turn vacuum cost accounting on or off */
PG_TRY();
{
ListCell *cur;
in_vacuum = true;
VacuumCostActive = (VacuumCostDelay > 0);
VacuumCostBalance = 0;
VacuumPageHit = 0;
VacuumPageMiss = 0;
VacuumPageDirty = 0;
/*
* Loop to process each selected relation.
*/
foreach(cur, relations)
{
Oid relid = lfirst_oid(cur);
if (options & VACOPT_VACUUM)
{
if (!vacuum_rel(relid, relation, options, params))
continue;
}
if (options & VACOPT_ANALYZE)
{
/*
* If using separate xacts, start one for analyze. Otherwise,
* we can use the outer transaction.
*/
if (use_own_xacts)
{
StartTransactionCommand();
/* functions in indexes may want a snapshot set */
PushActiveSnapshot(GetTransactionSnapshot());
}
analyze_rel(relid, relation, options, params,
va_cols, in_outer_xact, vac_strategy);
if (use_own_xacts)
{
PopActiveSnapshot();
CommitTransactionCommand();
}
}
}
}