Fix performance problems with autovacuum truncation in busy workloads.

In situations where there are over 8MB of empty pages at the end of
a table, the truncation work for trailing empty pages takes longer
than deadlock_timeout, and there is frequent access to the table by
processes other than autovacuum, there was a problem with the
autovacuum worker process being canceled by the deadlock checking
code. The truncation work done by autovacuum up that point was
lost, and the attempt tried again by a later autovacuum worker. The
attempts could continue indefinitely without making progress,
consuming resources and blocking other processes for up to
deadlock_timeout each time.

This patch has the autovacuum worker checking whether it is
blocking any other thread at 20ms intervals. If such a condition
develops, the autovacuum worker will persist the work it has done
so far, release its lock on the table, and sleep in 50ms intervals
for up to 5 seconds, hoping to be able to re-acquire the lock and
try again. If it is unable to get the lock in that time, it moves
on and a worker will try to continue later from the point this one
left off.

While this patch doesn't change the rules about when and what to
truncate, it does cause the truncation to occur sooner, with less
blocking, and with the consumption of fewer resources when there is
contention for the table's lock.

The only user-visible change other than improved performance is
that the table size during truncation may change incrementally
instead of just once.

This problem exists in all supported versions but is infrequently
reported, although some reports of performance problems when
autovacuum runs might be caused by this. Initial commit is just the
master branch, but this should probably be backpatched once the
build farm and general developer usage confirm that there are no
surprising effects.

Jan Wieck

1 files changed, 92 insertions, 0 deletions

diff --git a/src/backend/storage/lmgr/lock.c b/src/backend/storage/lmgr/lock.c
index 01834437467..ec4da20a0d4 100644
--- a/src/backend/storage/lmgr/lock.c
+++ b/src/backend/storage/lmgr/lock.c
@@ -538,6 +538,98 @@ ProcLockHashCode(const PROCLOCKTAG *proclocktag, uint32 hashcode)
 	return lockhash;
 }
 
+/*
+ * LockHasWaiters -- look up 'locktag' and check if releasing this
+ *		lock would wake up other processes waiting for it.
+ */
+bool
+LockHasWaiters(const LOCKTAG *locktag, LOCKMODE lockmode, bool sessionLock)
+{
+	LOCKMETHODID lockmethodid = locktag->locktag_lockmethodid;
+	LockMethod	lockMethodTable;
+	LOCALLOCKTAG localtag;
+	LOCALLOCK  *locallock;
+	LOCK	   *lock;
+	PROCLOCK   *proclock;
+	LWLockId	partitionLock;
+	bool		hasWaiters = false;
+
+	if (lockmethodid <= 0 || lockmethodid >= lengthof(LockMethods))
+		elog(ERROR, "unrecognized lock method: %d", lockmethodid);
+	lockMethodTable = LockMethods[lockmethodid];
+	if (lockmode <= 0 || lockmode > lockMethodTable->numLockModes)
+		elog(ERROR, "unrecognized lock mode: %d", lockmode);
+
+#ifdef LOCK_DEBUG
+	if (LOCK_DEBUG_ENABLED(locktag))
+		elog(LOG, "LockHasWaiters: lock [%u,%u] %s",
+			 locktag->locktag_field1, locktag->locktag_field2,
+			 lockMethodTable->lockModeNames[lockmode]);
+#endif
+
+	/*
+	 * Find the LOCALLOCK entry for this lock and lockmode
+	 */
+	MemSet(&localtag, 0, sizeof(localtag));		/* must clear padding */
+	localtag.lock = *locktag;
+	localtag.mode = lockmode;
+
+	locallock = (LOCALLOCK *) hash_search(LockMethodLocalHash,
+										  (void *) &localtag,
+										  HASH_FIND, NULL);
+
+	/*
+	 * let the caller print its own error message, too. Do not ereport(ERROR).
+	 */
+	if (!locallock || locallock->nLocks <= 0)
+	{
+		elog(WARNING, "you don't own a lock of type %s",
+			 lockMethodTable->lockModeNames[lockmode]);
+		return false;
+	}
+
+	/*
+	 * Check the shared lock table.
+	 */
+	partitionLock = LockHashPartitionLock(locallock->hashcode);
+
+	LWLockAcquire(partitionLock, LW_SHARED);
+
+	/*
+	 * We don't need to re-find the lock or proclock, since we kept their
+	 * addresses in the locallock table, and they couldn't have been removed
+	 * while we were holding a lock on them.
+	 */
+	lock = locallock->lock;
+	LOCK_PRINT("LockHasWaiters: found", lock, lockmode);
+	proclock = locallock->proclock;
+	PROCLOCK_PRINT("LockHasWaiters: found", proclock);
+
+	/*
+	 * Double-check that we are actually holding a lock of the type we want to
+	 * release.
+	 */
+	if (!(proclock->holdMask & LOCKBIT_ON(lockmode)))
+	{
+		PROCLOCK_PRINT("LockHasWaiters: WRONGTYPE", proclock);
+		LWLockRelease(partitionLock);
+		elog(WARNING, "you don't own a lock of type %s",
+			 lockMethodTable->lockModeNames[lockmode]);
+		RemoveLocalLock(locallock);
+		return false;
+	}
+
+	/*
+	 * Do the checking.
+	 */
+	if ((lockMethodTable->conflictTab[lockmode] & lock->waitMask) != 0)
+		hasWaiters = true;
+
+	LWLockRelease(partitionLock);
+
+	return hasWaiters;
+}
+
 
 /*
  * LockAcquire -- Check for lock conflicts, sleep if conflict found,