/*
* SubTransGetTopmostTransaction
*
* Returns the topmost transaction of the given transaction id.
*
* Because we cannot look back further than TransactionXmin, it is possible
* that this function will lie and return an intermediate subtransaction ID
* instead of the true topmost parent ID. This is OK, because in practice
* we only care about detecting whether the topmost parent is still running
* or is part of a current snapshot's list of still-running transactions.
* Therefore, any XID before TransactionXmin is as good as any other.
*/
TransactionId
SubTransGetTopmostTransaction(TransactionId xid)
{
TransactionId parentXid = xid,
previousXid = xid;
/* Can't ask about stuff that might not be around anymore */
Assert(TransactionIdFollowsOrEquals(xid, TransactionXmin));
while (TransactionIdIsValid(parentXid))
{
previousXid = parentXid;
if (TransactionIdPrecedes(parentXid, TransactionXmin))
break;
parentXid = SubTransGetParent(parentXid);
/*
* By convention the parent xid gets allocated first, so should always
* precede the child xid. Anything else points to a corrupted data
* structure that could lead to an infinite loop, so exit.
*/
if (!TransactionIdPrecedes(parentXid, previousXid))
elog(ERROR, "pg_subtrans contains invalid entry: xid %u points to parent xid %u",
previousXid, parentXid);
}
Assert(TransactionIdIsValid(previousXid));
return previousXid;
}
/*
* Get new XID. For global transaction is it previsly set by dtm_begin_transaction or dtm_join_transaction.
* Local transactions are using range of local Xids obtains from DTM.
*/
static TransactionId DtmGetNextXid()
{
TransactionId xid;
LWLockAcquire(dtm->xidLock, LW_EXCLUSIVE);
if (TransactionIdIsValid(DtmNextXid))
{
XTM_INFO("Use global XID %d\n", DtmNextXid);
xid = DtmNextXid;
if (TransactionIdPrecedesOrEquals(ShmemVariableCache->nextXid, xid))
{
/* Advance ShmemVariableCache->nextXid formward until new Xid */
while (TransactionIdPrecedes(ShmemVariableCache->nextXid, xid))
{
XTM_INFO("Extend CLOG for global transaction to %d\n", ShmemVariableCache->nextXid);
ExtendCLOG(ShmemVariableCache->nextXid);
ExtendCommitTs(ShmemVariableCache->nextXid);
ExtendSUBTRANS(ShmemVariableCache->nextXid);
TransactionIdAdvance(ShmemVariableCache->nextXid);
}
dtm->nReservedXids = 0;
}
}
else
{
if (dtm->nReservedXids == 0)
{
XTM_INFO("%d: reserve new XID range\n", getpid());
dtm->nReservedXids = ArbiterReserve(ShmemVariableCache->nextXid, DtmLocalXidReserve, &dtm->nextXid);
Assert(dtm->nReservedXids > 0);
Assert(TransactionIdFollowsOrEquals(dtm->nextXid, ShmemVariableCache->nextXid));
/* Advance ShmemVariableCache->nextXid formward until new Xid */
while (TransactionIdPrecedes(ShmemVariableCache->nextXid, dtm->nextXid))
{
XTM_INFO("Extend CLOG for local transaction to %d\n", ShmemVariableCache->nextXid);
ExtendCLOG(ShmemVariableCache->nextXid);
ExtendCommitTs(ShmemVariableCache->nextXid);
ExtendSUBTRANS(ShmemVariableCache->nextXid);
TransactionIdAdvance(ShmemVariableCache->nextXid);
}
}
Assert(ShmemVariableCache->nextXid == dtm->nextXid);
xid = dtm->nextXid++;
dtm->nReservedXids -= 1;
XTM_INFO("Obtain new local XID %d\n", xid);
}
LWLockRelease(dtm->xidLock);
return xid;
}
/*
* Compute the oldest xmin across all slots and store it in the ProcArray.
*/
void
ReplicationSlotsComputeRequiredXmin(bool already_locked)
{
int i;
TransactionId agg_xmin = InvalidTransactionId;
TransactionId agg_catalog_xmin = InvalidTransactionId;
Assert(ReplicationSlotCtl != NULL);
if (!already_locked)
LWLockAcquire(ReplicationSlotControlLock, LW_SHARED);
for (i = 0; i < max_replication_slots; i++)
{
ReplicationSlot *s = &ReplicationSlotCtl->replication_slots[i];
TransactionId effective_xmin;
TransactionId effective_catalog_xmin;
if (!s->in_use)
continue;
{
volatile ReplicationSlot *vslot = s;
SpinLockAcquire(&s->mutex);
effective_xmin = vslot->effective_xmin;
effective_catalog_xmin = vslot->effective_catalog_xmin;
SpinLockRelease(&s->mutex);
}
/* check the data xmin */
if (TransactionIdIsValid(effective_xmin) &&
(!TransactionIdIsValid(agg_xmin) ||
TransactionIdPrecedes(effective_xmin, agg_xmin)))
agg_xmin = effective_xmin;
/* check the catalog xmin */
if (TransactionIdIsValid(effective_catalog_xmin) &&
(!TransactionIdIsValid(agg_catalog_xmin) ||
TransactionIdPrecedes(effective_catalog_xmin, agg_catalog_xmin)))
agg_catalog_xmin = effective_catalog_xmin;
}
if (!already_locked)
LWLockRelease(ReplicationSlotControlLock);
ProcArraySetReplicationSlotXmin(agg_xmin, agg_catalog_xmin, already_locked);
}
/*
* do a TransactionId -> txid conversion for an XID near the given epoch
*/
static txid
convert_xid(TransactionId xid, const TxidEpoch *state)
{
#ifndef INT64_IS_BUSTED
uint64 epoch;
/* return special xid's as-is */
if (!TransactionIdIsNormal(xid))
return (txid) xid;
/* xid can be on either side when near wrap-around */
epoch = (uint64) state->epoch;
if (xid > state->last_xid &&
TransactionIdPrecedes(xid, state->last_xid))
epoch--;
else if (xid < state->last_xid &&
TransactionIdFollows(xid, state->last_xid))
epoch++;
return (epoch << 32) | xid;
#else /* INT64_IS_BUSTED */
/* we can't do anything with the epoch, so ignore it */
return (txid) xid & MAX_TXID;
#endif /* INT64_IS_BUSTED */
}
/*
* Move forwards the oldest commitTS value that can be consulted
*/
void
AdvanceOldestCommitTsXid(TransactionId oldestXact)
{
LWLockAcquire(CommitTsLock, LW_EXCLUSIVE);
if (ShmemVariableCache->oldestCommitTsXid != InvalidTransactionId &&
TransactionIdPrecedes(ShmemVariableCache->oldestCommitTsXid, oldestXact))
ShmemVariableCache->oldestCommitTsXid = oldestXact;
LWLockRelease(CommitTsLock);
}
/*
* TransactionIdDidCommit
* True iff transaction associated with the identifier did commit.
*
* Note:
* Assumes transaction identifier is valid.
*/
bool /* true if given transaction committed */
TransactionIdDidCommit(TransactionId transactionId)
{
XidStatus xidstatus;
xidstatus = TransactionLogFetch(transactionId);
/*
* If it's marked committed, it's committed.
*/
if (xidstatus == TRANSACTION_STATUS_COMMITTED)
#ifdef PGXC
{
syncGXID_GTM((GlobalTransactionId)transactionId);
#endif
return true;
#ifdef PGXC
}
#endif
/*
* If it's marked subcommitted, we have to check the parent recursively.
* However, if it's older than TransactionXmin, we can't look at
* pg_subtrans; instead assume that the parent crashed without cleaning up
* its children.
*
* Originally we Assert'ed that the result of SubTransGetParent was not
* zero. However with the introduction of prepared transactions, there can
* be a window just after database startup where we do not have complete
* knowledge in pg_subtrans of the transactions after TransactionXmin.
* StartupSUBTRANS() has ensured that any missing information will be
* zeroed. Since this case should not happen under normal conditions, it
* seems reasonable to emit a WARNING for it.
*/
if (xidstatus == TRANSACTION_STATUS_SUB_COMMITTED)
{
TransactionId parentXid;
if (TransactionIdPrecedes(transactionId, TransactionXmin))
return false;
parentXid = SubTransGetParent(transactionId);
if (!TransactionIdIsValid(parentXid))
{
elog(WARNING, "no pg_subtrans entry for subcommitted XID %u",
transactionId);
return false;
}
return TransactionIdDidCommit(parentXid);
}
/*
* It's not committed.
*/
return false;
}
/*
* GetOldestXmin -- returns oldest transaction that was running
* when any current transaction was started.
*
* If allDbs is TRUE then all backends are considered; if allDbs is FALSE
* then only backends running in my own database are considered.
*
* This is used by VACUUM to decide which deleted tuples must be preserved
* in a table. allDbs = TRUE is needed for shared relations, but allDbs =
* FALSE is sufficient for non-shared relations, since only backends in my
* own database could ever see the tuples in them.
*
* Note: we include the currently running xids in the set of considered xids.
* This ensures that if a just-started xact has not yet set its snapshot,
* when it does set the snapshot it cannot set xmin less than what we compute.
*/
TransactionId
GetOldestXmin(bool allDbs)
{
SISeg *segP = shmInvalBuffer;
ProcState *stateP = segP->procState;
TransactionId result;
int index;
result = GetCurrentTransactionId();
LWLockAcquire(SInvalLock, LW_SHARED);
for (index = 0; index < segP->lastBackend; index++)
{
SHMEM_OFFSET pOffset = stateP[index].procStruct;
if (pOffset != INVALID_OFFSET)
{
PGPROC *proc = (PGPROC *) MAKE_PTR(pOffset);
if (allDbs || proc->databaseId == MyDatabaseId)
{
/* Fetch xid just once - see GetNewTransactionId */
TransactionId xid = proc->xid;
if (TransactionIdIsNormal(xid))
{
if (TransactionIdPrecedes(xid, result))
result = xid;
xid = proc->xmin;
if (TransactionIdIsNormal(xid))
if (TransactionIdPrecedes(xid, result))
result = xid;
}
}
}
}
LWLockRelease(SInvalLock);
return result;
}
/*
* Set the limit values between which commit TS can be consulted.
*/
void
SetCommitTsLimit(TransactionId oldestXact, TransactionId newestXact)
{
/*
* Be careful not to overwrite values that are either further into the
* "future" or signal a disabled committs.
*/
LWLockAcquire(CommitTsLock, LW_EXCLUSIVE);
if (ShmemVariableCache->oldestCommitTs != InvalidTransactionId)
{
if (TransactionIdPrecedes(ShmemVariableCache->oldestCommitTs, oldestXact))
ShmemVariableCache->oldestCommitTs = oldestXact;
if (TransactionIdPrecedes(newestXact, ShmemVariableCache->newestCommitTs))
ShmemVariableCache->newestCommitTs = newestXact;
}
else
{
Assert(ShmemVariableCache->newestCommitTs == InvalidTransactionId);
}
LWLockRelease(CommitTsLock);
}
/* Record lowest soon-prunable XID */
static void
heap_prune_record_prunable(PruneState *prstate, TransactionId xid)
{
/*
* This should exactly match the PageSetPrunable macro. We can't store
* directly into the page header yet, so we update working state.
*/
Assert(TransactionIdIsNormal(xid));
if (!TransactionIdIsValid(prstate->new_prune_xid) ||
TransactionIdPrecedes(xid, prstate->new_prune_xid))
prstate->new_prune_xid = xid;
}
/*
* Comparison function for RegisteredSnapshots heap. Snapshots are ordered
* by xmin, so that the snapshot with smallest xmin is at the top.
*/
static int
xmin_cmp(const pairingheap_node *a, const pairingheap_node *b, void *arg)
{
const SnapshotData *asnap = pairingheap_const_container(SnapshotData, ph_node, a);
const SnapshotData *bsnap = pairingheap_const_container(SnapshotData, ph_node, b);
if (TransactionIdPrecedes(asnap->xmin, bsnap->xmin))
return 1;
else if (TransactionIdFollows(asnap->xmin, bsnap->xmin))
return -1;
else
return 0;
}
/*
* Decide which of two SUBTRANS page numbers is "older" for truncation purposes.
*
* We need to use comparison of TransactionIds here in order to do the right
* thing with wraparound XID arithmetic. However, if we are asked about
* page number zero, we don't want to hand InvalidTransactionId to
* TransactionIdPrecedes: it'll get weird about permanent xact IDs. So,
* offset both xids by FirstNormalTransactionId to avoid that.
*/
static bool
SubTransPagePrecedes(int page1, int page2)
{
TransactionId xid1;
TransactionId xid2;
xid1 = ((TransactionId) page1) * SUBTRANS_XACTS_PER_PAGE;
xid1 += FirstNormalTransactionId;
xid2 = ((TransactionId) page2) * SUBTRANS_XACTS_PER_PAGE;
xid2 += FirstNormalTransactionId;
return TransactionIdPrecedes(xid1, xid2);
}
/*
* Decide which of two CLOG page numbers is "older" for truncation purposes.
*
* We need to use comparison of TransactionIds here in order to do the right
* thing with wraparound XID arithmetic. However, if we are asked about
* page number zero, we don't want to hand InvalidTransactionId to
* TransactionIdPrecedes: it'll get weird about permanent xact IDs. So,
* offset both xids by FirstNormalTransactionId to avoid that.
*/
static bool
CommitTsPagePrecedes(int page1, int page2)
{
TransactionId xid1;
TransactionId xid2;
xid1 = ((TransactionId) page1) * COMMIT_TS_XACTS_PER_PAGE;
xid1 += FirstNormalTransactionId;
xid2 = ((TransactionId) page2) * COMMIT_TS_XACTS_PER_PAGE;
xid2 += FirstNormalTransactionId;
return TransactionIdPrecedes(xid1, xid2);
}
/*
* Decide which of two DistributedLog page numbers is "older" for
* truncation purposes.
*
* We need to use comparison of TransactionIds here in order to do the right
* thing with wraparound XID arithmetic. However, if we are asked about
* page number zero, we don't want to hand InvalidTransactionId to
* TransactionIdPrecedes: it'll get weird about permanent xact IDs. So,
* offset both xids by FirstNormalTransactionId to avoid that.
*/
static bool
DistributedLog_PagePrecedes(int page1, int page2)
{
TransactionId xid1;
TransactionId xid2;
xid1 = ((TransactionId) page1) * ENTRIES_PER_PAGE;
xid1 += FirstNormalTransactionId;
xid2 = ((TransactionId) page2) * ENTRIES_PER_PAGE;
xid2 += FirstNormalTransactionId;
return TransactionIdPrecedes(xid1, xid2);
}
/*
* Get oldest Xid visible by any active transaction (global or local)
* Take in account global Xmin received from DTMD
*/
static TransactionId DtmGetOldestXmin(Relation rel, bool ignoreVacuum)
{
TransactionId localXmin = PgGetOldestXmin(rel, ignoreVacuum);
TransactionId globalXmin = dtm->minXid;
XTM_INFO("XTM: DtmGetOldestXmin localXmin=%d, globalXmin=%d\n", localXmin, globalXmin);
if (TransactionIdIsValid(globalXmin))
{
globalXmin -= vacuum_defer_cleanup_age;
if (!TransactionIdIsNormal(globalXmin))
globalXmin = FirstNormalTransactionId;
if (TransactionIdPrecedes(globalXmin, localXmin))
localXmin = globalXmin;
XTM_INFO("XTM: DtmGetOldestXmin adjusted localXmin=%d, globalXmin=%d\n", localXmin, globalXmin);
}
return localXmin;
}
/*
* HeapTupleIsSurelyDead
*
* Determine whether a tuple is surely dead. We sometimes use this
* in lieu of HeapTupleSatisifesVacuum when the tuple has just been
* tested by HeapTupleSatisfiesMVCC and, therefore, any hint bits that
* can be set should already be set. We assume that if no hint bits
* either for xmin or xmax, the transaction is still running. This is
* therefore faster than HeapTupleSatisfiesVacuum, because we don't
* consult CLOG (and also because we don't need to give an exact answer,
* just whether or not the tuple is surely dead).
*/
bool
HeapTupleIsSurelyDead(HeapTuple htup, TransactionId OldestXmin)
{
HeapTupleHeader tuple = htup->t_data;
Assert(ItemPointerIsValid(&htup->t_self));
Assert(htup->t_tableOid != InvalidOid);
/*
* If the inserting transaction is marked invalid, then it aborted, and
* the tuple is definitely dead. If it's marked neither committed nor
* invalid, then we assume it's still alive (since the presumption is that
* all relevant hint bits were just set moments ago).
*/
if (!HeapTupleHeaderXminCommitted(tuple))
return HeapTupleHeaderXminInvalid(tuple) ? true : false;
/*
* If the inserting transaction committed, but any deleting transaction
* aborted, the tuple is still alive.
*/
if (tuple->t_infomask & HEAP_XMAX_INVALID)
return false;
/*
* If the XMAX is just a lock, the tuple is still alive.
*/
if (HEAP_XMAX_IS_LOCKED_ONLY(tuple->t_infomask))
return false;
/*
* If the Xmax is a MultiXact, it might be dead or alive, but we cannot
* know without checking pg_multixact.
*/
if (tuple->t_infomask & HEAP_XMAX_IS_MULTI)
return false;
/* If deleter isn't known to have committed, assume it's still running. */
if (!(tuple->t_infomask & HEAP_XMAX_COMMITTED))
return false;
/* Deleter committed, so tuple is dead if the XID is old enough. */
return TransactionIdPrecedes(HeapTupleHeaderGetRawXmax(tuple), OldestXmin);
}
/* adapted from Postgres' txid.c#convert_xid function */
uint64 convert_xid(TransactionId xid) {
TxidEpoch state;
uint64 epoch;
GetNextXidAndEpoch(&state.last_xid, &state.epoch);
/* return special xid's as-is */
if (!TransactionIdIsNormal(xid))
return (uint64) xid;
/* xid can be on either side when near wrap-around */
epoch = (uint64) state.epoch;
if (xid > state.last_xid && TransactionIdPrecedes(xid, state.last_xid))
epoch--;
else if (xid < state.last_xid && TransactionIdFollows(xid, state.last_xid))
epoch++;
return (epoch << 32) | xid;
}
/*
* SnapshotResetXmin
*
* If there are no more snapshots, we can reset our PGXACT->xmin to InvalidXid.
* Note we can do this without locking because we assume that storing an Xid
* is atomic.
*
* Even if there are some remaining snapshots, we may be able to advance our
* PGXACT->xmin to some degree. This typically happens when a portal is
* dropped. For efficiency, we only consider recomputing PGXACT->xmin when
* the active snapshot stack is empty.
*/
static void
SnapshotResetXmin(void)
{
Snapshot minSnapshot;
if (ActiveSnapshot != NULL)
return;
if (pairingheap_is_empty(&RegisteredSnapshots))
{
MyPgXact->xmin = InvalidTransactionId;
return;
}
minSnapshot = pairingheap_container(SnapshotData, ph_node,
pairingheap_first(&RegisteredSnapshots));
if (TransactionIdPrecedes(MyPgXact->xmin, minSnapshot->xmin))
MyPgXact->xmin = minSnapshot->xmin;
}
/*
* do a TransactionId -> txid conversion for an XID near the given epoch
*/
static txid
convert_xid(TransactionId xid, const TxidEpoch *state)
{
uint64 epoch;
/* return special xid's as-is */
if (!TransactionIdIsNormal(xid))
return (txid) xid;
/* xid can be on either side when near wrap-around */
epoch = (uint64) state->epoch;
if (xid > state->last_xid &&
TransactionIdPrecedes(xid, state->last_xid))
epoch--;
else if (xid < state->last_xid &&
TransactionIdFollows(xid, state->last_xid))
epoch++;
return (epoch << 32) | xid;
}
/*
* TransactionIdDidAbort
* True iff transaction associated with the identifier did abort.
*
* Note:
* Assumes transaction identifier is valid.
*/
bool /* true if given transaction aborted */
TransactionIdDidAbort(TransactionId transactionId)
{
XidStatus xidstatus;
xidstatus = TransactionLogFetch(transactionId);
/*
* If it's marked aborted, it's aborted.
*/
if (xidstatus == TRANSACTION_STATUS_ABORTED)
return true;
/*
* If it's marked subcommitted, we have to check the parent recursively.
* However, if it's older than TransactionXmin, we can't look at
* pg_subtrans; instead assume that the parent crashed without cleaning up
* its children.
*/
if (xidstatus == TRANSACTION_STATUS_SUB_COMMITTED)
{
TransactionId parentXid;
if (TransactionIdPrecedes(transactionId, TransactionXmin))
return true;
parentXid = SubTransGetParent(transactionId);
if (!TransactionIdIsValid(parentXid))
{
/* see notes in TransactionIdDidCommit */
elog(WARNING, "no pg_subtrans entry for subcommitted XID %u",
transactionId);
return true;
}
return TransactionIdDidAbort(parentXid);
}
/*
* It's not aborted.
*/
return false;
}
/*
* SubTransGetTopmostTransaction
*
* Returns the topmost transaction of the given transaction id.
*
* Because we cannot look back further than TransactionXmin, it is possible
* that this function will lie and return an intermediate subtransaction ID
* instead of the true topmost parent ID. This is OK, because in practice
* we only care about detecting whether the topmost parent is still running
* or is part of a current snapshot's list of still-running transactions.
* Therefore, any XID before TransactionXmin is as good as any other.
*/
TransactionId
SubTransGetTopmostTransaction(TransactionId xid)
{
TransactionId parentXid = xid,
previousXid = xid;
/* Can't ask about stuff that might not be around anymore */
Assert(TransactionIdFollowsOrEquals(xid, TransactionXmin));
while (TransactionIdIsValid(parentXid))
{
previousXid = parentXid;
if (TransactionIdPrecedes(parentXid, TransactionXmin))
break;
parentXid = SubTransGetParent(parentXid);
}
Assert(TransactionIdIsValid(previousXid));
return previousXid;
}
/*
* TransactionIdLatest --- get latest XID among a main xact and its children
*/
TransactionId
TransactionIdLatest(TransactionId mainxid,
int nxids, const TransactionId *xids)
{
TransactionId result;
/*
* In practice it is highly likely that the xids[] array is sorted, and so
* we could save some cycles by just taking the last child XID, but this
* probably isn't so performance-critical that it's worth depending on
* that assumption. But just to show we're not totally stupid, scan the
* array back-to-front to avoid useless assignments.
*/
result = mainxid;
while (--nxids >= 0)
{
if (TransactionIdPrecedes(result, xids[nxids]))
result = xids[nxids];
}
return result;
}
/*
* Check whether a tuple is all-visible relative to a given OldestXmin value.
* The buffer should contain the tuple and should be locked and pinned.
*/
static bool
tuple_all_visible(HeapTuple tup, TransactionId OldestXmin, Buffer buffer)
{
HTSV_Result state;
TransactionId xmin;
state = HeapTupleSatisfiesVacuum(tup, OldestXmin, buffer);
if (state != HEAPTUPLE_LIVE)
return false; /* all-visible implies live */
/*
* Neither lazy_scan_heap nor heap_page_is_all_visible will mark a page
* all-visible unless every tuple is hinted committed. However, those hint
* bits could be lost after a crash, so we can't be certain that they'll
* be set here. So just check the xmin.
*/
xmin = HeapTupleHeaderGetXmin(tup->t_data);
if (!TransactionIdPrecedes(xmin, OldestXmin))
return false; /* xmin not old enough for all to see */
return true;
}
/*
* StandbyReleaseLocksMany
* Release standby locks held by XIDs < removeXid
*
* If keepPreparedXacts is true, keep prepared transactions even if
* they're older than removeXid
*/
static void
StandbyReleaseLocksMany(TransactionId removeXid, bool keepPreparedXacts)
{
ListCell *cell,
*prev,
*next;
LOCKTAG locktag;
/*
* Release all matching locks.
*/
prev = NULL;
for (cell = list_head(RecoveryLockList); cell; cell = next)
{
xl_standby_lock *lock = (xl_standby_lock *) lfirst(cell);
next = lnext(cell);
if (!TransactionIdIsValid(removeXid) || TransactionIdPrecedes(lock->xid, removeXid))
{
if (keepPreparedXacts && StandbyTransactionIdIsPrepared(lock->xid))
continue;
elog(trace_recovery(DEBUG4),
"releasing recovery lock: xid %u db %u rel %u",
lock->xid, lock->dbOid, lock->relOid);
SET_LOCKTAG_RELATION(locktag, lock->dbOid, lock->relOid);
if (!LockRelease(&locktag, AccessExclusiveLock, true))
elog(LOG,
"RecoveryLockList contains entry for lock no longer recorded by lock manager: xid %u database %u relation %u",
lock->xid, lock->dbOid, lock->relOid);
RecoveryLockList = list_delete_cell(RecoveryLockList, cell, prev);
pfree(lock);
}
else
prev = cell;
}
}
/*
* _bt_page_recyclable() -- Is an existing page recyclable?
*
* This exists to make sure _bt_getbuf and btvacuumscan have the same
* policy about whether a page is safe to re-use.
*/
bool
_bt_page_recyclable(Page page)
{
BTPageOpaque opaque;
/*
* It's possible to find an all-zeroes page in an index --- for example, a
* backend might successfully extend the relation one page and then crash
* before it is able to make a WAL entry for adding the page. If we find a
* zeroed page then reclaim it.
*/
if (PageIsNew(page))
return true;
/*
* Otherwise, recycle if deleted and too old to have any processes
* interested in it.
*/
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
if (P_ISDELETED(opaque) &&
TransactionIdPrecedes(opaque->btpo.xact, RecentGlobalXmin))
return true;
return false;
}
/*
* lazy_scan_heap() -- scan an open heap relation
*
* This routine sets commit status bits, builds lists of dead tuples
* and pages with free space, and calculates statistics on the number
* of live tuples in the heap. When done, or when we run low on space
* for dead-tuple TIDs, invoke vacuuming of indexes and heap.
*
* If there are no indexes then we just vacuum each dirty page as we
* process it, since there's no point in gathering many tuples.
*/
static void
lazy_scan_heap(Relation onerel, LVRelStats *vacrelstats,
Relation *Irel, int nindexes, bool scan_all)
{
BlockNumber nblocks,
blkno;
HeapTupleData tuple;
char *relname;
BlockNumber empty_pages,
vacuumed_pages;
double num_tuples,
tups_vacuumed,
nkeep,
nunused;
IndexBulkDeleteResult **indstats;
int i;
PGRUsage ru0;
Buffer vmbuffer = InvalidBuffer;
BlockNumber next_not_all_visible_block;
bool skipping_all_visible_blocks;
pg_rusage_init(&ru0);
relname = RelationGetRelationName(onerel);
ereport(elevel,
(errmsg("vacuuming \"%s.%s\"",
get_namespace_name(RelationGetNamespace(onerel)),
relname)));
empty_pages = vacuumed_pages = 0;
num_tuples = tups_vacuumed = nkeep = nunused = 0;
indstats = (IndexBulkDeleteResult **)
palloc0(nindexes * sizeof(IndexBulkDeleteResult *));
nblocks = RelationGetNumberOfBlocks(onerel);
vacrelstats->rel_pages = nblocks;
vacrelstats->scanned_pages = 0;
vacrelstats->nonempty_pages = 0;
vacrelstats->latestRemovedXid = InvalidTransactionId;
lazy_space_alloc(vacrelstats, nblocks);
/*
* We want to skip pages that don't require vacuuming according to the
* visibility map, but only when we can skip at least SKIP_PAGES_THRESHOLD
* consecutive pages. Since we're reading sequentially, the OS should be
* doing readahead for us, so there's no gain in skipping a page now and
* then; that's likely to disable readahead and so be counterproductive.
* Also, skipping even a single page means that we can't update
* relfrozenxid, so we only want to do it if we can skip a goodly number
* of pages.
*
* Before entering the main loop, establish the invariant that
* next_not_all_visible_block is the next block number >= blkno that's not
* all-visible according to the visibility map, or nblocks if there's no
* such block. Also, we set up the skipping_all_visible_blocks flag,
* which is needed because we need hysteresis in the decision: once we've
* started skipping blocks, we may as well skip everything up to the next
* not-all-visible block.
*
* Note: if scan_all is true, we won't actually skip any pages; but we
* maintain next_not_all_visible_block anyway, so as to set up the
* all_visible_according_to_vm flag correctly for each page.
*/
for (next_not_all_visible_block = 0;
next_not_all_visible_block < nblocks;
next_not_all_visible_block++)
{
if (!visibilitymap_test(onerel, next_not_all_visible_block, &vmbuffer))
break;
vacuum_delay_point();
}
if (next_not_all_visible_block >= SKIP_PAGES_THRESHOLD)
skipping_all_visible_blocks = true;
else
skipping_all_visible_blocks = false;
for (blkno = 0; blkno < nblocks; blkno++)
{
Buffer buf;
Page page;
OffsetNumber offnum,
maxoff;
bool tupgone,
hastup;
int prev_dead_count;
OffsetNumber frozen[MaxOffsetNumber];
int nfrozen;
Size freespace;
bool all_visible_according_to_vm;
bool all_visible;
bool has_dead_tuples;
if (blkno == next_not_all_visible_block)
{
/* Time to advance next_not_all_visible_block */
for (next_not_all_visible_block++;
next_not_all_visible_block < nblocks;
next_not_all_visible_block++)
{
if (!visibilitymap_test(onerel, next_not_all_visible_block,
&vmbuffer))
break;
vacuum_delay_point();
}
/*
* We know we can't skip the current block. But set up
* skipping_all_visible_blocks to do the right thing at the
* following blocks.
*/
if (next_not_all_visible_block - blkno > SKIP_PAGES_THRESHOLD)
skipping_all_visible_blocks = true;
else
skipping_all_visible_blocks = false;
all_visible_according_to_vm = false;
}
else
{
/* Current block is all-visible */
if (skipping_all_visible_blocks && !scan_all)
continue;
all_visible_according_to_vm = true;
}
vacuum_delay_point();
vacrelstats->scanned_pages++;
/*
* If we are close to overrunning the available space for dead-tuple
* TIDs, pause and do a cycle of vacuuming before we tackle this page.
*/
if ((vacrelstats->max_dead_tuples - vacrelstats->num_dead_tuples) < MaxHeapTuplesPerPage &&
vacrelstats->num_dead_tuples > 0)
{
/* Log cleanup info before we touch indexes */
vacuum_log_cleanup_info(onerel, vacrelstats);
/* Remove index entries */
for (i = 0; i < nindexes; i++)
lazy_vacuum_index(Irel[i],
&indstats[i],
vacrelstats);
/* Remove tuples from heap */
lazy_vacuum_heap(onerel, vacrelstats);
/*
* Forget the now-vacuumed tuples, and press on, but be careful
* not to reset latestRemovedXid since we want that value to be
* valid.
*/
vacrelstats->num_dead_tuples = 0;
vacrelstats->num_index_scans++;
}
buf = ReadBufferExtended(onerel, MAIN_FORKNUM, blkno,
RBM_NORMAL, vac_strategy);
/* We need buffer cleanup lock so that we can prune HOT chains. */
LockBufferForCleanup(buf);
page = BufferGetPage(buf);
if (PageIsNew(page))
{
/*
* An all-zeroes page could be left over if a backend extends the
* relation but crashes before initializing the page. Reclaim such
* pages for use.
*
* We have to be careful here because we could be looking at a
* page that someone has just added to the relation and not yet
* been able to initialize (see RelationGetBufferForTuple). To
* protect against that, release the buffer lock, grab the
* relation extension lock momentarily, and re-lock the buffer. If
* the page is still uninitialized by then, it must be left over
* from a crashed backend, and we can initialize it.
*
* We don't really need the relation lock when this is a new or
* temp relation, but it's probably not worth the code space to
* check that, since this surely isn't a critical path.
*
* Note: the comparable code in vacuum.c need not worry because
* it's got exclusive lock on the whole relation.
*/
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
LockRelationForExtension(onerel, ExclusiveLock);
UnlockRelationForExtension(onerel, ExclusiveLock);
LockBufferForCleanup(buf);
if (PageIsNew(page))
{
ereport(WARNING,
(errmsg("relation \"%s\" page %u is uninitialized --- fixing",
relname, blkno)));
PageInit(page, BufferGetPageSize(buf), 0);
empty_pages++;
}
freespace = PageGetHeapFreeSpace(page);
MarkBufferDirty(buf);
UnlockReleaseBuffer(buf);
RecordPageWithFreeSpace(onerel, blkno, freespace);
continue;
}
if (PageIsEmpty(page))
{
empty_pages++;
freespace = PageGetHeapFreeSpace(page);
if (!PageIsAllVisible(page))
{
PageSetAllVisible(page);
SetBufferCommitInfoNeedsSave(buf);
}
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
/* Update the visibility map */
if (!all_visible_according_to_vm)
{
visibilitymap_pin(onerel, blkno, &vmbuffer);
LockBuffer(buf, BUFFER_LOCK_SHARE);
if (PageIsAllVisible(page))
visibilitymap_set(onerel, blkno, PageGetLSN(page), &vmbuffer);
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
}
ReleaseBuffer(buf);
RecordPageWithFreeSpace(onerel, blkno, freespace);
continue;
}
/*
* Prune all HOT-update chains in this page.
*
* We count tuples removed by the pruning step as removed by VACUUM.
*/
tups_vacuumed += heap_page_prune(onerel, buf, OldestXmin, false,
&vacrelstats->latestRemovedXid);
/*
* Now scan the page to collect vacuumable items and check for tuples
* requiring freezing.
*/
all_visible = true;
has_dead_tuples = false;
nfrozen = 0;
hastup = false;
prev_dead_count = vacrelstats->num_dead_tuples;
maxoff = PageGetMaxOffsetNumber(page);
for (offnum = FirstOffsetNumber;
offnum <= maxoff;
offnum = OffsetNumberNext(offnum))
{
ItemId itemid;
itemid = PageGetItemId(page, offnum);
/* Unused items require no processing, but we count 'em */
if (!ItemIdIsUsed(itemid))
{
nunused += 1;
continue;
}
/* Redirect items mustn't be touched */
if (ItemIdIsRedirected(itemid))
{
hastup = true; /* this page won't be truncatable */
continue;
}
ItemPointerSet(&(tuple.t_self), blkno, offnum);
/*
* DEAD item pointers are to be vacuumed normally; but we don't
* count them in tups_vacuumed, else we'd be double-counting (at
* least in the common case where heap_page_prune() just freed up
* a non-HOT tuple).
*/
if (ItemIdIsDead(itemid))
{
lazy_record_dead_tuple(vacrelstats, &(tuple.t_self));
all_visible = false;
continue;
}
Assert(ItemIdIsNormal(itemid));
tuple.t_data = (HeapTupleHeader) PageGetItem(page, itemid);
tuple.t_len = ItemIdGetLength(itemid);
tupgone = false;
switch (HeapTupleSatisfiesVacuum(tuple.t_data, OldestXmin, buf))
{
case HEAPTUPLE_DEAD:
/*
* Ordinarily, DEAD tuples would have been removed by
* heap_page_prune(), but it's possible that the tuple
* state changed since heap_page_prune() looked. In
* particular an INSERT_IN_PROGRESS tuple could have
* changed to DEAD if the inserter aborted. So this
* cannot be considered an error condition.
*
* If the tuple is HOT-updated then it must only be
* removed by a prune operation; so we keep it just as if
* it were RECENTLY_DEAD. Also, if it's a heap-only
* tuple, we choose to keep it, because it'll be a lot
* cheaper to get rid of it in the next pruning pass than
* to treat it like an indexed tuple.
*/
if (HeapTupleIsHotUpdated(&tuple) ||
HeapTupleIsHeapOnly(&tuple))
nkeep += 1;
else
tupgone = true; /* we can delete the tuple */
all_visible = false;
break;
case HEAPTUPLE_LIVE:
/* Tuple is good --- but let's do some validity checks */
if (onerel->rd_rel->relhasoids &&
!OidIsValid(HeapTupleGetOid(&tuple)))
elog(WARNING, "relation \"%s\" TID %u/%u: OID is invalid",
relname, blkno, offnum);
/*
* Is the tuple definitely visible to all transactions?
*
* NB: Like with per-tuple hint bits, we can't set the
* PD_ALL_VISIBLE flag if the inserter committed
* asynchronously. See SetHintBits for more info. Check
* that the HEAP_XMIN_COMMITTED hint bit is set because of
* that.
*/
if (all_visible)
{
TransactionId xmin;
if (!(tuple.t_data->t_infomask & HEAP_XMIN_COMMITTED))
{
all_visible = false;
break;
}
/*
* The inserter definitely committed. But is it old
* enough that everyone sees it as committed?
*/
xmin = HeapTupleHeaderGetXmin(tuple.t_data);
if (!TransactionIdPrecedes(xmin, OldestXmin))
{
all_visible = false;
break;
}
}
break;
case HEAPTUPLE_RECENTLY_DEAD:
/*
* If tuple is recently deleted then we must not remove it
* from relation.
*/
nkeep += 1;
all_visible = false;
break;
case HEAPTUPLE_INSERT_IN_PROGRESS:
/* This is an expected case during concurrent vacuum */
all_visible = false;
break;
case HEAPTUPLE_DELETE_IN_PROGRESS:
/* This is an expected case during concurrent vacuum */
all_visible = false;
break;
default:
elog(ERROR, "unexpected HeapTupleSatisfiesVacuum result");
break;
}
if (tupgone)
{
lazy_record_dead_tuple(vacrelstats, &(tuple.t_self));
HeapTupleHeaderAdvanceLatestRemovedXid(tuple.t_data,
&vacrelstats->latestRemovedXid);
tups_vacuumed += 1;
has_dead_tuples = true;
}
else
{
num_tuples += 1;
hastup = true;
/*
* Each non-removable tuple must be checked to see if it needs
* freezing. Note we already have exclusive buffer lock.
*/
if (heap_freeze_tuple(tuple.t_data, FreezeLimit,
InvalidBuffer))
frozen[nfrozen++] = offnum;
}
} /* scan along page */
/*
* If we froze any tuples, mark the buffer dirty, and write a WAL
* record recording the changes. We must log the changes to be
* crash-safe against future truncation of CLOG.
*/
if (nfrozen > 0)
{
MarkBufferDirty(buf);
if (RelationNeedsWAL(onerel))
{
XLogRecPtr recptr;
recptr = log_heap_freeze(onerel, buf, FreezeLimit,
frozen, nfrozen);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
}
}
/*
* If there are no indexes then we can vacuum the page right now
* instead of doing a second scan.
*/
if (nindexes == 0 &&
vacrelstats->num_dead_tuples > 0)
{
/* Remove tuples from heap */
lazy_vacuum_page(onerel, blkno, buf, 0, vacrelstats);
/*
* Forget the now-vacuumed tuples, and press on, but be careful
* not to reset latestRemovedXid since we want that value to be
* valid.
*/
vacrelstats->num_dead_tuples = 0;
vacuumed_pages++;
}
freespace = PageGetHeapFreeSpace(page);
/* Update the all-visible flag on the page */
if (!PageIsAllVisible(page) && all_visible)
{
PageSetAllVisible(page);
SetBufferCommitInfoNeedsSave(buf);
}
/*
* It's possible for the value returned by GetOldestXmin() to move
* backwards, so it's not wrong for us to see tuples that appear to
* not be visible to everyone yet, while PD_ALL_VISIBLE is already
* set. The real safe xmin value never moves backwards, but
* GetOldestXmin() is conservative and sometimes returns a value
* that's unnecessarily small, so if we see that contradiction it just
* means that the tuples that we think are not visible to everyone yet
* actually are, and the PD_ALL_VISIBLE flag is correct.
*
* There should never be dead tuples on a page with PD_ALL_VISIBLE
* set, however.
*/
else if (PageIsAllVisible(page) && has_dead_tuples)
{
elog(WARNING, "page containing dead tuples is marked as all-visible in relation \"%s\" page %u",
relname, blkno);
PageClearAllVisible(page);
SetBufferCommitInfoNeedsSave(buf);
/*
* Normally, we would drop the lock on the heap page before
* updating the visibility map, but since this case shouldn't
* happen anyway, don't worry about that.
*/
visibilitymap_clear(onerel, blkno);
}
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
/* Update the visibility map */
if (!all_visible_according_to_vm && all_visible)
{
visibilitymap_pin(onerel, blkno, &vmbuffer);
LockBuffer(buf, BUFFER_LOCK_SHARE);
if (PageIsAllVisible(page))
visibilitymap_set(onerel, blkno, PageGetLSN(page), &vmbuffer);
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
}
ReleaseBuffer(buf);
/* Remember the location of the last page with nonremovable tuples */
if (hastup)
vacrelstats->nonempty_pages = blkno + 1;
/*
* If we remembered any tuples for deletion, then the page will be
* visited again by lazy_vacuum_heap, which will compute and record
* its post-compaction free space. If not, then we're done with this
* page, so remember its free space as-is. (This path will always be
* taken if there are no indexes.)
*/
if (vacrelstats->num_dead_tuples == prev_dead_count)
RecordPageWithFreeSpace(onerel, blkno, freespace);
}
/* save stats for use later */
vacrelstats->scanned_tuples = num_tuples;
vacrelstats->tuples_deleted = tups_vacuumed;
/* now we can compute the new value for pg_class.reltuples */
vacrelstats->new_rel_tuples = vac_estimate_reltuples(onerel, false,
nblocks,
vacrelstats->scanned_pages,
num_tuples);
/* If any tuples need to be deleted, perform final vacuum cycle */
/* XXX put a threshold on min number of tuples here? */
if (vacrelstats->num_dead_tuples > 0)
{
/* Log cleanup info before we touch indexes */
vacuum_log_cleanup_info(onerel, vacrelstats);
/* Remove index entries */
for (i = 0; i < nindexes; i++)
lazy_vacuum_index(Irel[i],
&indstats[i],
vacrelstats);
/* Remove tuples from heap */
lazy_vacuum_heap(onerel, vacrelstats);
vacrelstats->num_index_scans++;
}
/* Release the pin on the visibility map page */
if (BufferIsValid(vmbuffer))
{
ReleaseBuffer(vmbuffer);
vmbuffer = InvalidBuffer;
}
/* Do post-vacuum cleanup and statistics update for each index */
for (i = 0; i < nindexes; i++)
lazy_cleanup_index(Irel[i], indstats[i], vacrelstats);
/* If no indexes, make log report that lazy_vacuum_heap would've made */
if (vacuumed_pages)
ereport(elevel,
(errmsg("\"%s\": removed %.0f row versions in %u pages",
RelationGetRelationName(onerel),
tups_vacuumed, vacuumed_pages)));
ereport(elevel,
(errmsg("\"%s\": found %.0f removable, %.0f nonremovable row versions in %u out of %u pages",
RelationGetRelationName(onerel),
tups_vacuumed, num_tuples,
vacrelstats->scanned_pages, nblocks),
errdetail("%.0f dead row versions cannot be removed yet.\n"
"There were %.0f unused item pointers.\n"
"%u pages are entirely empty.\n"
"%s.",
nkeep,
nunused,
empty_pages,
pg_rusage_show(&ru0))));
}
/*
* Interrogate the commit timestamp of a transaction.
*
* The return value indicates whether a commit timestamp record was found for
* the given xid. The timestamp value is returned in *ts (which may not be
* null), and the origin node for the Xid is returned in *nodeid, if it's not
* null.
*/
bool
TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
RepOriginId *nodeid)
{
int pageno = TransactionIdToCTsPage(xid);
int entryno = TransactionIdToCTsEntry(xid);
int slotno;
CommitTimestampEntry entry;
TransactionId oldestCommitTsXid;
TransactionId newestCommitTsXid;
if (!TransactionIdIsValid(xid))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("cannot retrieve commit timestamp for transaction %u", xid)));
else if (!TransactionIdIsNormal(xid))
{
/* frozen and bootstrap xids are always committed far in the past */
*ts = 0;
if (nodeid)
*nodeid = 0;
return false;
}
LWLockAcquire(CommitTsLock, LW_SHARED);
/* Error if module not enabled */
if (!commitTsShared->commitTsActive)
error_commit_ts_disabled();
/*
* If we're asked for the cached value, return that. Otherwise, fall
* through to read from SLRU.
*/
if (commitTsShared->xidLastCommit == xid)
{
*ts = commitTsShared->dataLastCommit.time;
if (nodeid)
*nodeid = commitTsShared->dataLastCommit.nodeid;
LWLockRelease(CommitTsLock);
return *ts != 0;
}
oldestCommitTsXid = ShmemVariableCache->oldestCommitTsXid;
newestCommitTsXid = ShmemVariableCache->newestCommitTsXid;
/* neither is invalid, or both are */
Assert(TransactionIdIsValid(oldestCommitTsXid) == TransactionIdIsValid(newestCommitTsXid));
LWLockRelease(CommitTsLock);
/*
* Return empty if the requested value is outside our valid range.
*/
if (!TransactionIdIsValid(oldestCommitTsXid) ||
TransactionIdPrecedes(xid, oldestCommitTsXid) ||
TransactionIdPrecedes(newestCommitTsXid, xid))
{
*ts = 0;
if (nodeid)
*nodeid = InvalidRepOriginId;
return false;
}
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
CommitTsCtl->shared->page_buffer[slotno] +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
*ts = entry.time;
if (nodeid)
*nodeid = entry.nodeid;
LWLockRelease(CommitTsControlLock);
return *ts != 0;
}
/*
* TransactionTreeSetCommitTsData
*
* Record the final commit timestamp of transaction entries in the commit log
* for a transaction and its subtransaction tree, as efficiently as possible.
*
* xid is the top level transaction id.
*
* subxids is an array of xids of length nsubxids, representing subtransactions
* in the tree of xid. In various cases nsubxids may be zero.
* The reason why tracking just the parent xid commit timestamp is not enough
* is that the subtrans SLRU does not stay valid across crashes (it's not
* permanent) so we need to keep the information about them here. If the
* subtrans implementation changes in the future, we might want to revisit the
* decision of storing timestamp info for each subxid.
*
* The write_xlog parameter tells us whether to include an XLog record of this
* or not. Normally, this is called from transaction commit routines (both
* normal and prepared) and the information will be stored in the transaction
* commit XLog record, and so they should pass "false" for this. The XLog redo
* code should use "false" here as well. Other callers probably want to pass
* true, so that the given values persist in case of crashes.
*/
void
TransactionTreeSetCommitTsData(TransactionId xid, int nsubxids,
TransactionId *subxids, TimestampTz timestamp,
RepOriginId nodeid, bool write_xlog)
{
int i;
TransactionId headxid;
TransactionId newestXact;
/*
* No-op if the module is not active.
*
* An unlocked read here is fine, because in a standby (the only place
* where the flag can change in flight) this routine is only called by the
* recovery process, which is also the only process which can change the
* flag.
*/
if (!commitTsShared->commitTsActive)
return;
/*
* Comply with the WAL-before-data rule: if caller specified it wants this
* value to be recorded in WAL, do so before touching the data.
*/
if (write_xlog)
WriteSetTimestampXlogRec(xid, nsubxids, subxids, timestamp, nodeid);
/*
* Figure out the latest Xid in this batch: either the last subxid if
* there's any, otherwise the parent xid.
*/
if (nsubxids > 0)
newestXact = subxids[nsubxids - 1];
else
newestXact = xid;
/*
* We split the xids to set the timestamp to in groups belonging to the
* same SLRU page; the first element in each such set is its head. The
* first group has the main XID as the head; subsequent sets use the first
* subxid not on the previous page as head. This way, we only have to
* lock/modify each SLRU page once.
*/
for (i = 0, headxid = xid;;)
{
int pageno = TransactionIdToCTsPage(headxid);
int j;
for (j = i; j < nsubxids; j++)
{
if (TransactionIdToCTsPage(subxids[j]) != pageno)
break;
}
/* subxids[i..j] are on the same page as the head */
SetXidCommitTsInPage(headxid, j - i, subxids + i, timestamp, nodeid,
pageno);
/* if we wrote out all subxids, we're done. */
if (j + 1 >= nsubxids)
break;
/*
* Set the new head and skip over it, as well as over the subxids we
* just wrote.
*/
headxid = subxids[j];
i += j - i + 1;
}
/* update the cached value in shared memory */
LWLockAcquire(CommitTsLock, LW_EXCLUSIVE);
commitTsShared->xidLastCommit = xid;
commitTsShared->dataLastCommit.time = timestamp;
commitTsShared->dataLastCommit.nodeid = nodeid;
/* and move forwards our endpoint, if needed */
if (TransactionIdPrecedes(ShmemVariableCache->newestCommitTsXid, newestXact))
ShmemVariableCache->newestCommitTsXid = newestXact;
LWLockRelease(CommitTsLock);
}
/*
* get_relation_info -
* Retrieves catalog information for a given relation.
*
* Given the Oid of the relation, return the following info into fields
* of the RelOptInfo struct:
*
* min_attr lowest valid AttrNumber
* max_attr highest valid AttrNumber
* indexlist list of IndexOptInfos for relation's indexes
* pages number of pages
* tuples number of tuples
*
* Also, initialize the attr_needed[] and attr_widths[] arrays. In most
* cases these are left as zeroes, but sometimes we need to compute attr
* widths here, and we may as well cache the results for costsize.c.
*
* If inhparent is true, all we need to do is set up the attr arrays:
* the RelOptInfo actually represents the appendrel formed by an inheritance
* tree, and so the parent rel's physical size and index information isn't
* important for it.
*/
void
get_relation_info(PlannerInfo *root, Oid relationObjectId, bool inhparent,
RelOptInfo *rel)
{
Index varno = rel->relid;
Relation relation;
bool hasindex;
List *indexinfos = NIL;
/*
* We need not lock the relation since it was already locked, either by
* the rewriter or when expand_inherited_rtentry() added it to the query's
* rangetable.
*/
relation = heap_open(relationObjectId, NoLock);
rel->min_attr = FirstLowInvalidHeapAttributeNumber + 1;
rel->max_attr = RelationGetNumberOfAttributes(relation);
rel->reltablespace = RelationGetForm(relation)->reltablespace;
Assert(rel->max_attr >= rel->min_attr);
rel->attr_needed = (Relids *)
palloc0((rel->max_attr - rel->min_attr + 1) * sizeof(Relids));
rel->attr_widths = (int32 *)
palloc0((rel->max_attr - rel->min_attr + 1) * sizeof(int32));
/*
* Estimate relation size --- unless it's an inheritance parent, in which
* case the size will be computed later in set_append_rel_pathlist, and we
* must leave it zero for now to avoid bollixing the total_table_pages
* calculation.
*/
if (!inhparent)
estimate_rel_size(relation, rel->attr_widths - rel->min_attr,
&rel->pages, &rel->tuples);
/*
* Make list of indexes. Ignore indexes on system catalogs if told to.
* Don't bother with indexes for an inheritance parent, either.
*/
if (inhparent ||
(IgnoreSystemIndexes && IsSystemClass(relation->rd_rel)))
hasindex = false;
else
hasindex = relation->rd_rel->relhasindex;
if (hasindex)
{
List *indexoidlist;
ListCell *l;
LOCKMODE lmode;
indexoidlist = RelationGetIndexList(relation);
/*
* For each index, we get the same type of lock that the executor will
* need, and do not release it. This saves a couple of trips to the
* shared lock manager while not creating any real loss of
* concurrency, because no schema changes could be happening on the
* index while we hold lock on the parent rel, and neither lock type
* blocks any other kind of index operation.
*/
if (rel->relid == root->parse->resultRelation)
lmode = RowExclusiveLock;
else
lmode = AccessShareLock;
foreach(l, indexoidlist)
{
Oid indexoid = lfirst_oid(l);
Relation indexRelation;
Form_pg_index index;
IndexOptInfo *info;
int ncolumns;
int i;
/*
* Extract info from the relation descriptor for the index.
*/
indexRelation = index_open(indexoid, lmode);
index = indexRelation->rd_index;
/*
* Ignore invalid indexes, since they can't safely be used for
* queries. Note that this is OK because the data structure we
* are constructing is only used by the planner --- the executor
* still needs to insert into "invalid" indexes!
*/
if (!index->indisvalid)
{
index_close(indexRelation, NoLock);
continue;
}
/*
* If the index is valid, but cannot yet be used, ignore it; but
* mark the plan we are generating as transient. See
* src/backend/access/heap/README.HOT for discussion.
*/
if (index->indcheckxmin &&
!TransactionIdPrecedes(HeapTupleHeaderGetXmin(indexRelation->rd_indextuple->t_data),
TransactionXmin))
{
root->glob->transientPlan = true;
index_close(indexRelation, NoLock);
continue;
}
info = makeNode(IndexOptInfo);
info->indexoid = index->indexrelid;
info->reltablespace =
RelationGetForm(indexRelation)->reltablespace;
info->rel = rel;
info->ncolumns = ncolumns = index->indnatts;
/*
* Allocate per-column info arrays. To save a few palloc cycles
* we allocate all the Oid-type arrays in one request. Note that
* the opfamily array needs an extra, terminating zero at the end.
* We pre-zero the ordering info in case the index is unordered.
*/
info->indexkeys = (int *) palloc(sizeof(int) * ncolumns);
info->opfamily = (Oid *) palloc0(sizeof(Oid) * (4 * ncolumns + 1));
info->opcintype = info->opfamily + (ncolumns + 1);
info->fwdsortop = info->opcintype + ncolumns;
info->revsortop = info->fwdsortop + ncolumns;
info->nulls_first = (bool *) palloc0(sizeof(bool) * ncolumns);
for (i = 0; i < ncolumns; i++)
{
info->indexkeys[i] = index->indkey.values[i];
info->opfamily[i] = indexRelation->rd_opfamily[i];
info->opcintype[i] = indexRelation->rd_opcintype[i];
}
info->relam = indexRelation->rd_rel->relam;
info->amcostestimate = indexRelation->rd_am->amcostestimate;
info->amoptionalkey = indexRelation->rd_am->amoptionalkey;
info->amsearchnulls = indexRelation->rd_am->amsearchnulls;
info->amhasgettuple = OidIsValid(indexRelation->rd_am->amgettuple);
info->amhasgetbitmap = OidIsValid(indexRelation->rd_am->amgetbitmap);
/*
* Fetch the ordering operators associated with the index, if any.
* We expect that all ordering-capable indexes use btree's
* strategy numbers for the ordering operators.
*/
if (indexRelation->rd_am->amcanorder)
{
int nstrat = indexRelation->rd_am->amstrategies;
for (i = 0; i < ncolumns; i++)
{
int16 opt = indexRelation->rd_indoption[i];
int fwdstrat;
int revstrat;
if (opt & INDOPTION_DESC)
{
fwdstrat = BTGreaterStrategyNumber;
revstrat = BTLessStrategyNumber;
}
else
{
fwdstrat = BTLessStrategyNumber;
revstrat = BTGreaterStrategyNumber;
}
/*
* Index AM must have a fixed set of strategies for it to
* make sense to specify amcanorder, so we need not allow
* the case amstrategies == 0.
*/
if (fwdstrat > 0)
{
Assert(fwdstrat <= nstrat);
info->fwdsortop[i] = indexRelation->rd_operator[i * nstrat + fwdstrat - 1];
}
if (revstrat > 0)
{
Assert(revstrat <= nstrat);
info->revsortop[i] = indexRelation->rd_operator[i * nstrat + revstrat - 1];
}
info->nulls_first[i] = (opt & INDOPTION_NULLS_FIRST) != 0;
}
}
/*
* Fetch the index expressions and predicate, if any. We must
* modify the copies we obtain from the relcache to have the
* correct varno for the parent relation, so that they match up
* correctly against qual clauses.
*/
info->indexprs = RelationGetIndexExpressions(indexRelation);
info->indpred = RelationGetIndexPredicate(indexRelation);
if (info->indexprs && varno != 1)
ChangeVarNodes((Node *) info->indexprs, 1, varno, 0);
if (info->indpred && varno != 1)
ChangeVarNodes((Node *) info->indpred, 1, varno, 0);
info->predOK = false; /* set later in indxpath.c */
info->unique = index->indisunique;
/*
* Estimate the index size. If it's not a partial index, we lock
* the number-of-tuples estimate to equal the parent table; if it
* is partial then we have to use the same methods as we would for
* a table, except we can be sure that the index is not larger
* than the table.
*/
if (info->indpred == NIL)
{
info->pages = RelationGetNumberOfBlocks(indexRelation);
info->tuples = rel->tuples;
}
else
{
estimate_rel_size(indexRelation, NULL,
&info->pages, &info->tuples);
if (info->tuples > rel->tuples)
info->tuples = rel->tuples;
}
index_close(indexRelation, NoLock);
indexinfos = lcons(info, indexinfos);
}
list_free(indexoidlist);
}
/*
* get_relation_info -
* Retrieves catalog information for a given relation.
*
* Given the Oid of the relation, return the following info into fields
* of the RelOptInfo struct:
*
* min_attr lowest valid AttrNumber
* max_attr highest valid AttrNumber
* indexlist list of IndexOptInfos for relation's indexes
* pages number of pages
* tuples number of tuples
*
* Also, initialize the attr_needed[] and attr_widths[] arrays. In most
* cases these are left as zeroes, but sometimes we need to compute attr
* widths here, and we may as well cache the results for costsize.c.
*
* If inhparent is true, all we need to do is set up the attr arrays:
* the RelOptInfo actually represents the appendrel formed by an inheritance
* tree, and so the parent rel's physical size and index information isn't
* important for it.
*/
void
get_relation_info(PlannerInfo *root, Oid relationObjectId, bool inhparent,
RelOptInfo *rel)
{
Index varno = rel->relid;
Relation relation;
bool hasindex;
List *indexinfos = NIL;
/*
* We need not lock the relation since it was already locked, either by
* the rewriter or when expand_inherited_rtentry() added it to the query's
* rangetable.
*/
relation = heap_open(relationObjectId, NoLock);
/* Temporary and unlogged relations are inaccessible during recovery. */
if (!RelationNeedsWAL(relation) && RecoveryInProgress())
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary or unlogged relations during recovery")));
rel->min_attr = FirstLowInvalidHeapAttributeNumber + 1;
rel->max_attr = RelationGetNumberOfAttributes(relation);
rel->reltablespace = RelationGetForm(relation)->reltablespace;
Assert(rel->max_attr >= rel->min_attr);
rel->attr_needed = (Relids *)
palloc0((rel->max_attr - rel->min_attr + 1) * sizeof(Relids));
rel->attr_widths = (int32 *)
palloc0((rel->max_attr - rel->min_attr + 1) * sizeof(int32));
/*
* Estimate relation size --- unless it's an inheritance parent, in which
* case the size will be computed later in set_append_rel_pathlist, and we
* must leave it zero for now to avoid bollixing the total_table_pages
* calculation.
*/
if (!inhparent)
estimate_rel_size(relation, rel->attr_widths - rel->min_attr,
&rel->pages, &rel->tuples, &rel->allvisfrac);
/*
* Make list of indexes. Ignore indexes on system catalogs if told to.
* Don't bother with indexes for an inheritance parent, either.
*/
if (inhparent ||
(IgnoreSystemIndexes && IsSystemClass(relation->rd_rel)))
hasindex = false;
else
hasindex = relation->rd_rel->relhasindex;
if (hasindex)
{
List *indexoidlist;
ListCell *l;
LOCKMODE lmode;
indexoidlist = RelationGetIndexList(relation);
/*
* For each index, we get the same type of lock that the executor will
* need, and do not release it. This saves a couple of trips to the
* shared lock manager while not creating any real loss of
* concurrency, because no schema changes could be happening on the
* index while we hold lock on the parent rel, and neither lock type
* blocks any other kind of index operation.
*/
if (rel->relid == root->parse->resultRelation)
lmode = RowExclusiveLock;
else
lmode = AccessShareLock;
foreach(l, indexoidlist)
{
Oid indexoid = lfirst_oid(l);
Relation indexRelation;
Form_pg_index index;
IndexOptInfo *info;
int ncolumns;
int i;
/*
* Extract info from the relation descriptor for the index.
*/
indexRelation = index_open(indexoid, lmode);
index = indexRelation->rd_index;
/*
* Ignore invalid indexes, since they can't safely be used for
* queries. Note that this is OK because the data structure we
* are constructing is only used by the planner --- the executor
* still needs to insert into "invalid" indexes!
*/
if (!index->indisvalid)
{
index_close(indexRelation, NoLock);
continue;
}
/*
* If the index is valid, but cannot yet be used, ignore it; but
* mark the plan we are generating as transient. See
* src/backend/access/heap/README.HOT for discussion.
*/
if (index->indcheckxmin &&
!TransactionIdPrecedes(HeapTupleHeaderGetXmin(indexRelation->rd_indextuple->t_data),
TransactionXmin))
{
root->glob->transientPlan = true;
index_close(indexRelation, NoLock);
continue;
}
info = makeNode(IndexOptInfo);
info->indexoid = index->indexrelid;
info->reltablespace =
RelationGetForm(indexRelation)->reltablespace;
info->rel = rel;
info->ncolumns = ncolumns = index->indnatts;
info->indexkeys = (int *) palloc(sizeof(int) * ncolumns);
info->indexcollations = (Oid *) palloc(sizeof(Oid) * ncolumns);
info->opfamily = (Oid *) palloc(sizeof(Oid) * ncolumns);
info->opcintype = (Oid *) palloc(sizeof(Oid) * ncolumns);
for (i = 0; i < ncolumns; i++)
{
info->indexkeys[i] = index->indkey.values[i];
info->indexcollations[i] = indexRelation->rd_indcollation[i];
info->opfamily[i] = indexRelation->rd_opfamily[i];
info->opcintype[i] = indexRelation->rd_opcintype[i];
}
info->relam = indexRelation->rd_rel->relam;
info->amcostestimate = indexRelation->rd_am->amcostestimate;
info->canreturn = index_can_return(indexRelation);
info->amcanorderbyop = indexRelation->rd_am->amcanorderbyop;
info->amoptionalkey = indexRelation->rd_am->amoptionalkey;
info->amsearcharray = indexRelation->rd_am->amsearcharray;
info->amsearchnulls = indexRelation->rd_am->amsearchnulls;
info->amhasgettuple = OidIsValid(indexRelation->rd_am->amgettuple);
info->amhasgetbitmap = OidIsValid(indexRelation->rd_am->amgetbitmap);
/*
* Fetch the ordering information for the index, if any.
*/
if (info->relam == BTREE_AM_OID)
{
/*
* If it's a btree index, we can use its opfamily OIDs
* directly as the sort ordering opfamily OIDs.
*/
Assert(indexRelation->rd_am->amcanorder);
info->sortopfamily = info->opfamily;
info->reverse_sort = (bool *) palloc(sizeof(bool) * ncolumns);
info->nulls_first = (bool *) palloc(sizeof(bool) * ncolumns);
for (i = 0; i < ncolumns; i++)
{
int16 opt = indexRelation->rd_indoption[i];
info->reverse_sort[i] = (opt & INDOPTION_DESC) != 0;
info->nulls_first[i] = (opt & INDOPTION_NULLS_FIRST) != 0;
}
}
else if (indexRelation->rd_am->amcanorder)
{
/*
* Otherwise, identify the corresponding btree opfamilies by
* trying to map this index's "<" operators into btree. Since
* "<" uniquely defines the behavior of a sort order, this is
* a sufficient test.
*
* XXX This method is rather slow and also requires the
* undesirable assumption that the other index AM numbers its
* strategies the same as btree. It'd be better to have a way
* to explicitly declare the corresponding btree opfamily for
* each opfamily of the other index type. But given the lack
* of current or foreseeable amcanorder index types, it's not
* worth expending more effort on now.
*/
info->sortopfamily = (Oid *) palloc(sizeof(Oid) * ncolumns);
info->reverse_sort = (bool *) palloc(sizeof(bool) * ncolumns);
info->nulls_first = (bool *) palloc(sizeof(bool) * ncolumns);
for (i = 0; i < ncolumns; i++)
{
int16 opt = indexRelation->rd_indoption[i];
Oid ltopr;
Oid btopfamily;
Oid btopcintype;
int16 btstrategy;
info->reverse_sort[i] = (opt & INDOPTION_DESC) != 0;
info->nulls_first[i] = (opt & INDOPTION_NULLS_FIRST) != 0;
ltopr = get_opfamily_member(info->opfamily[i],
info->opcintype[i],
info->opcintype[i],
BTLessStrategyNumber);
if (OidIsValid(ltopr) &&
get_ordering_op_properties(ltopr,
&btopfamily,
&btopcintype,
&btstrategy) &&
btopcintype == info->opcintype[i] &&
btstrategy == BTLessStrategyNumber)
{
/* Successful mapping */
info->sortopfamily[i] = btopfamily;
}
else
{
/* Fail ... quietly treat index as unordered */
info->sortopfamily = NULL;
info->reverse_sort = NULL;
info->nulls_first = NULL;
break;
}
}
}
else
{
info->sortopfamily = NULL;
info->reverse_sort = NULL;
info->nulls_first = NULL;
}
/*
* Fetch the index expressions and predicate, if any. We must
* modify the copies we obtain from the relcache to have the
* correct varno for the parent relation, so that they match up
* correctly against qual clauses.
*/
info->indexprs = RelationGetIndexExpressions(indexRelation);
info->indpred = RelationGetIndexPredicate(indexRelation);
if (info->indexprs && varno != 1)
ChangeVarNodes((Node *) info->indexprs, 1, varno, 0);
if (info->indpred && varno != 1)
ChangeVarNodes((Node *) info->indpred, 1, varno, 0);
/* Build targetlist using the completed indexprs data */
info->indextlist = build_index_tlist(root, info, relation);
info->predOK = false; /* set later in indxpath.c */
info->unique = index->indisunique;
info->immediate = index->indimmediate;
info->hypothetical = false;
/*
* Estimate the index size. If it's not a partial index, we lock
* the number-of-tuples estimate to equal the parent table; if it
* is partial then we have to use the same methods as we would for
* a table, except we can be sure that the index is not larger
* than the table.
*/
if (info->indpred == NIL)
{
info->pages = RelationGetNumberOfBlocks(indexRelation);
info->tuples = rel->tuples;
}
else
{
double allvisfrac; /* dummy */
estimate_rel_size(indexRelation, NULL,
&info->pages, &info->tuples, &allvisfrac);
if (info->tuples > rel->tuples)
info->tuples = rel->tuples;
}
index_close(indexRelation, NoLock);
indexinfos = lcons(info, indexinfos);
}
list_free(indexoidlist);
}
/*
* Hot Standby feedback
*/
static void
ProcessStandbyHSFeedbackMessage(void)
{
StandbyHSFeedbackMessage reply;
TransactionId newxmin = InvalidTransactionId;
pq_copymsgbytes(&reply_message, (char *) &reply, sizeof(StandbyHSFeedbackMessage));
elog(DEBUG2, "hot standby feedback xmin %u epoch %u",
reply.xmin,
reply.epoch);
/*
* Update the WalSender's proc xmin to allow it to be visible to
* snapshots. This will hold back the removal of dead rows and thereby
* prevent the generation of cleanup conflicts on the standby server.
*/
if (TransactionIdIsValid(reply.xmin))
{
TransactionId nextXid;
uint32 nextEpoch;
bool epochOK = false;
GetNextXidAndEpoch(&nextXid, &nextEpoch);
/*
* Epoch of oldestXmin should be same as standby or if the counter has
* wrapped, then one less than reply.
*/
if (reply.xmin <= nextXid)
{
if (reply.epoch == nextEpoch)
epochOK = true;
}
else
{
if (nextEpoch > 0 && reply.epoch == nextEpoch - 1)
epochOK = true;
}
/*
* Feedback from standby must not go backwards, nor should it go
* forwards further than our most recent xid.
*/
if (epochOK && TransactionIdPrecedesOrEquals(reply.xmin, nextXid))
{
if (!TransactionIdIsValid(MyProc->xmin))
{
TransactionId oldestXmin = GetOldestXmin(true, true);
if (TransactionIdPrecedes(oldestXmin, reply.xmin))
newxmin = reply.xmin;
else
newxmin = oldestXmin;
}
else
{
if (TransactionIdPrecedes(MyProc->xmin, reply.xmin))
newxmin = reply.xmin;
else
newxmin = MyProc->xmin; /* stay the same */
}
}
}
/*
* Grab the ProcArrayLock to set xmin, or invalidate for bad reply
*/
if (MyProc->xmin != newxmin)
{
LWLockAcquire(ProcArrayLock, LW_SHARED);
MyProc->xmin = newxmin;
LWLockRelease(ProcArrayLock);
}
}
