Re-implement EvalPlanQual processing to improve its performance and eliminate

a lot of strange behaviors that occurred in join cases.  We now identify the
"current" row for every joined relation in UPDATE, DELETE, and SELECT FOR
UPDATE/SHARE queries.  If an EvalPlanQual recheck is necessary, we jam the
appropriate row into each scan node in the rechecking plan, forcing it to emit
only that one row.  The former behavior could rescan the whole of each joined
relation for each recheck, which was terrible for performance, and what's much
worse could result in duplicated output tuples.

Also, the original implementation of EvalPlanQual could not re-use the recheck
execution tree --- it had to go through a full executor init and shutdown for
every row to be tested.  To avoid this overhead, I've associated a special
runtime Param with each LockRows or ModifyTable plan node, and arranged to
make every scan node below such a node depend on that Param.  Thus, by
signaling a change in that Param, the EPQ machinery can just rescan the
already-built test plan.

This patch also adds a prohibition on set-returning functions in the
targetlist of SELECT FOR UPDATE/SHARE.  This is needed to avoid the
duplicate-output-tuple problem.  It seems fairly reasonable since the
other restrictions on SELECT FOR UPDATE are meant to ensure that there
is a unique correspondence between source tuples and result tuples,
which an output SRF destroys as much as anything else does.

1 files changed, 56 insertions, 12 deletions

diff --git a/src/backend/executor/nodeModifyTable.c b/src/backend/executor/nodeModifyTable.c
index a9fd8c4974f..3f1f9c093ee 100644
--- a/src/backend/executor/nodeModifyTable.c
+++ b/src/backend/executor/nodeModifyTable.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/executor/nodeModifyTable.c,v 1.1 2009/10/10 01:43:47 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/executor/nodeModifyTable.c,v 1.2 2009/10/26 02:26:31 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -273,7 +273,7 @@ ExecInsert(TupleTableSlot *slot,
 static TupleTableSlot *
 ExecDelete(ItemPointer tupleid,
 		   TupleTableSlot *planSlot,
-		   PlanState *subplanstate,
+		   EPQState *epqstate,
 		   EState *estate)
 {
 	ResultRelInfo *resultRelInfo;
@@ -294,7 +294,7 @@ ExecDelete(ItemPointer tupleid,
 	{
 		bool		dodelete;
 
-		dodelete = ExecBRDeleteTriggers(estate, subplanstate, resultRelInfo,
+		dodelete = ExecBRDeleteTriggers(estate, epqstate, resultRelInfo,
 										tupleid);
 
 		if (!dodelete)			/* "do nothing" */
@@ -329,13 +329,14 @@ ldelete:;
 				ereport(ERROR,
 						(errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
 						 errmsg("could not serialize access due to concurrent update")));
-			else if (!ItemPointerEquals(tupleid, &update_ctid))
+			if (!ItemPointerEquals(tupleid, &update_ctid))
 			{
 				TupleTableSlot *epqslot;
 
 				epqslot = EvalPlanQual(estate,
+									   epqstate,
+									   resultRelationDesc,
 									   resultRelInfo->ri_RangeTableIndex,
-									   subplanstate,
 									   &update_ctid,
 									   update_xmax);
 				if (!TupIsNull(epqslot))
@@ -416,7 +417,7 @@ static TupleTableSlot *
 ExecUpdate(ItemPointer tupleid,
 		   TupleTableSlot *slot,
 		   TupleTableSlot *planSlot,
-		   PlanState *subplanstate,
+		   EPQState *epqstate,
 		   EState *estate)
 {
 	HeapTuple	tuple;
@@ -451,7 +452,7 @@ ExecUpdate(ItemPointer tupleid,
 	{
 		HeapTuple	newtuple;
 
-		newtuple = ExecBRUpdateTriggers(estate, subplanstate, resultRelInfo,
+		newtuple = ExecBRUpdateTriggers(estate, epqstate, resultRelInfo,
 										tupleid, tuple);
 
 		if (newtuple == NULL)	/* "do nothing" */
@@ -515,13 +516,14 @@ lreplace:;
 				ereport(ERROR,
 						(errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
 						 errmsg("could not serialize access due to concurrent update")));
-			else if (!ItemPointerEquals(tupleid, &update_ctid))
+			if (!ItemPointerEquals(tupleid, &update_ctid))
 			{
 				TupleTableSlot *epqslot;
 
 				epqslot = EvalPlanQual(estate,
+									   epqstate,
+									   resultRelationDesc,
 									   resultRelInfo->ri_RangeTableIndex,
-									   subplanstate,
 									   &update_ctid,
 									   update_xmax);
 				if (!TupIsNull(epqslot))
@@ -685,12 +687,14 @@ ExecModifyTable(ModifyTableState *node)
 				estate->es_result_relation_info++;
 				subplanstate = node->mt_plans[node->mt_whichplan];
 				junkfilter = estate->es_result_relation_info->ri_junkFilter;
+				EvalPlanQualSetPlan(&node->mt_epqstate, subplanstate->plan);
 				continue;
 			}
 			else
 				break;
 		}
 
+		EvalPlanQualSetSlot(&node->mt_epqstate, planSlot);
 		slot = planSlot;
 
 		if (junkfilter != NULL)
@@ -728,11 +732,11 @@ ExecModifyTable(ModifyTableState *node)
 				break;
 			case CMD_UPDATE:
 				slot = ExecUpdate(tupleid, slot, planSlot,
-								  subplanstate, estate);
+								  &node->mt_epqstate, estate);
 				break;
 			case CMD_DELETE:
 				slot = ExecDelete(tupleid, planSlot,
-								  subplanstate, estate);
+								  &node->mt_epqstate, estate);
 				break;
 			default:
 				elog(ERROR, "unknown operation");
@@ -785,7 +789,7 @@ ExecInitModifyTable(ModifyTable *node, EState *estate, int eflags)
 	 * a subplan tree to EvalPlanQual, instead.  Use a runtime test not just
 	 * Assert because this condition is easy to miss in testing ...
 	 */
-	if (estate->es_evTuple != NULL)
+	if (estate->es_epqTuple != NULL)
 		elog(ERROR, "ModifyTable should not be called during EvalPlanQual");
 
 	/*
@@ -799,6 +803,8 @@ ExecInitModifyTable(ModifyTable *node, EState *estate, int eflags)
 	mtstate->mt_plans = (PlanState **) palloc0(sizeof(PlanState *) * nplans);
 	mtstate->mt_nplans = nplans;
 	mtstate->operation = operation;
+	/* set up epqstate with dummy subplan pointer for the moment */
+	EvalPlanQualInit(&mtstate->mt_epqstate, estate, NULL, node->epqParam);
 	mtstate->fireBSTriggers = true;
 
 	/* For the moment, assume our targets are exactly the global result rels */
@@ -823,6 +829,7 @@ ExecInitModifyTable(ModifyTable *node, EState *estate, int eflags)
 	/* select first subplan */
 	mtstate->mt_whichplan = 0;
 	subplan = (Plan *) linitial(node->plans);
+	EvalPlanQualSetPlan(&mtstate->mt_epqstate, subplan);
 
 	/*
 	 * Initialize RETURNING projections if needed.
@@ -879,6 +886,38 @@ ExecInitModifyTable(ModifyTable *node, EState *estate, int eflags)
 	}
 
 	/*
+	 * If we have any secondary relations in an UPDATE or DELETE, they need
+	 * to be treated like non-locked relations in SELECT FOR UPDATE, ie,
+	 * the EvalPlanQual mechanism needs to be told about them.  Locate
+	 * the relevant ExecRowMarks.
+	 */
+	foreach(l, node->rowMarks)
+	{
+		PlanRowMark *rc = (PlanRowMark *) lfirst(l);
+		ExecRowMark *erm = NULL;
+		ListCell   *lce;
+
+		Assert(IsA(rc, PlanRowMark));
+
+		/* ignore "parent" rowmarks; they are irrelevant at runtime */
+		if (rc->isParent)
+			continue;
+
+		foreach(lce, estate->es_rowMarks)
+		{
+			erm = (ExecRowMark *) lfirst(lce);
+			if (erm->rti == rc->rti)
+				break;
+			erm = NULL;
+		}
+		if (erm == NULL)
+			elog(ERROR, "failed to find ExecRowMark for PlanRowMark %u",
+				 rc->rti);
+
+		EvalPlanQualAddRowMark(&mtstate->mt_epqstate, erm);
+	}
+
+	/*
 	 * Initialize the junk filter(s) if needed.  INSERT queries need a filter
 	 * if there are any junk attrs in the tlist.  UPDATE and DELETE
 	 * always need a filter, since there's always a junk 'ctid' attribute
@@ -988,6 +1027,11 @@ ExecEndModifyTable(ModifyTableState *node)
 	ExecClearTuple(node->ps.ps_ResultTupleSlot);
 
 	/*
+	 * Terminate EPQ execution if active
+	 */
+	EvalPlanQualEnd(&node->mt_epqstate);
+
+	/*
 	 * shut down subplans
 	 */
 	for (i=0; i<node->mt_nplans; i++)