Compare Xmin to previous Xmax when locking an update chain

alvherre · alvherre · commit 2a4b6eed0916 · 2013-11-28T11:54:25.000-03:00
Not doing so causes us to traverse an update chain that has been broken
by concurrent page pruning.  All other code that traverses update chains
uses this check as one of the cases in which to stop iterating, so
replicate it here too.  Failure to do so leads to erroneous CLOG,
subtrans or multixact lookups.

Per discussion following the bug report by J Smith in
CADFUPgc5bmtv-yg9znxV-vcfkb+JPRqs7m2OesQXaM_4Z1JpdQ@mail.gmail.com
as diagnosed by Andres Freund.
diff --git a/src/backend/access/heap/heapam.c b/src/backend/access/heap/heapam.c
@@ -4797,6 +4797,7 @@ heap_lock_updated_tuple_rec(Relation rel, ItemPointer tid, TransactionId xid,
 				old_infomask;
 	TransactionId xmax,
 				new_xmax;
+	TransactionId priorXmax = InvalidTransactionId;
 
 	ItemPointerCopy(tid, &tupid);
 
@@ -4822,6 +4823,18 @@ heap_lock_updated_tuple_rec(Relation rel, ItemPointer tid, TransactionId xid,
 		CHECK_FOR_INTERRUPTS();
 		LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
 
+		/*
+		 * Check the tuple XMIN against prior XMAX, if any.  If we reached
+		 * the end of the chain, we're done, so return success.
+		 */
+		if (TransactionIdIsValid(priorXmax) &&
+			!TransactionIdEquals(HeapTupleHeaderGetXmin(mytup.t_data),
+								 priorXmax))
+		{
+			UnlockReleaseBuffer(buf);
+			return HeapTupleMayBeUpdated;
+		}
+
 		old_infomask = mytup.t_data->t_infomask;
 		xmax = HeapTupleHeaderGetRawXmax(mytup.t_data);
 
@@ -4922,6 +4935,7 @@ heap_lock_updated_tuple_rec(Relation rel, ItemPointer tid, TransactionId xid,
 		}
 
 		/* tail recursion */
+		priorXmax = HeapTupleHeaderGetUpdateXid(mytup.t_data);
 		ItemPointerCopy(&(mytup.t_data->t_ctid), &tupid);
 		UnlockReleaseBuffer(buf);
 	}
