Fix segfault in EXPLAIN with LEFT JOIN and correlated subqueries (#8548)

src/backend/distributed/planner/distributed_planner.c

@@ -83,7 +83,7 @@ int PlannerLevel = 0;
 
 static bool ListContainsDistributedTableRTE(List *rangeTableList,
 											bool *maybeHasForeignDistributedTable);
-static bool PlanContainsDistributedSubPlanRTE(List *subPlanList);
+static bool PlanContainsDistributedSubPlanRTE(DistributedPlanningContext *planContext);
 static PlannedStmt * CreateDistributedPlannedStmt(DistributedPlanningContext *
 												  planContext);
 static PlannedStmt * InlineCtesAndCreateDistributedPlannedStmt(uint64 planId,
@@ -436,8 +436,9 @@ ListContainsDistributedTableRTE(List *rangeTableList,
 
 
 /*
- * PlanContainsDistributedSubPlanRTE checks whether any of the subplans in the given
- * subPlanList is a Read Intermediate Result function scan.
+ * PlanContainsDistributedSubPlanRTE checks whether any of the subplans in the
+ * plan context's PlannedStmt->subplans list is a Read Intermediate Result
+ * function scan.
  *
  * It is used by the check after standard_planner() to determine whether the plan
  * still requires distributed planning; in addition to checking the range table for
@@ -446,14 +447,26 @@ ListContainsDistributedTableRTE(List *rangeTableList,
  * that distributed planning is required.
  */
 static bool
-PlanContainsDistributedSubPlanRTE(List *subPlanList)
+PlanContainsDistributedSubPlanRTE(DistributedPlanningContext *planContext)
 {
+	/*
+	 * We iterate over planContext->plan->subplans, which is PostgreSQL's
+	 * PlannedStmt->subplans list. PostgreSQL's setrefs.c (set_plan_references)
+	 * resolves AlternativeSubPlan nodes by picking one alternative and setting
+	 * the discarded subplan entries to NULL. We must therefore skip NULL entries.
+	 */
+	List *subPlanList = planContext->plan->subplans;
 	ListCell *subPlanCell = NULL;
 
 	foreach(subPlanCell, subPlanList)
 	{
 		Node *planRoot = (Node *) lfirst(subPlanCell);
 
+		if (planRoot == NULL)
+		{
+			continue;
+		}
+
 		if (!IsA(planRoot, FunctionScan))
 		{
 			continue;
@@ -2975,7 +2988,7 @@ CheckPostPlanDistribution(DistributedPlanningContext *planContext, bool
 			/* ..or a distributed subplan */
 			planHasDistribution = planHasDistribution ||
 								  PlanContainsDistributedSubPlanRTE(
-				planContext->plan->subplans);
+				planContext);
 
 			/*
 			 * The plan has a distributed relation, so we know for sure that

src/test/regress/expected/subquery_in_where.out

@@ -1414,6 +1414,71 @@ WHERE true OR NOT EXISTS (SELECT 1 FROM t1);
                1
 (1 row)
 
+-- Test crash fix for issue #8548
+-- A query with LEFT JOIN to a distributed table and correlated subqueries
+-- could crash because PostgreSQL's setrefs.c sets subplan list entries to
+-- NULL when AlternativeSubPlan resolution discards unused alternatives.
+-- PlanContainsDistributedSubPlanRTE must skip NULL entries.
+CREATE TABLE t4 (vkey integer, pkey integer, c30 integer, c31 integer, c32 text);
+CREATE TABLE t5 (vkey integer, pkey integer, c33 text, c34 integer, c35 integer,
+                 c36 timestamp without time zone);
+CREATE TABLE t2 (vkey integer, pkey integer, c15 numeric, c16 timestamp without time zone,
+                 c17 text, c18 text, c19 timestamp without time zone,
+                 c20 timestamp without time zone, c21 integer);
+CREATE TABLE t22 (vkey integer, pkey integer, c37 numeric, c38 text, c39 numeric,
+                  c40 numeric, c41 numeric, c42 integer,
+                  c43 timestamp without time zone, c44 numeric,
+                  colocated_key numeric);
+SELECT create_distributed_table('t22', 'colocated_key');
+ create_distributed_table
+---------------------------------------------------------------------
+
+(1 row)
+
+-- This query should not crash (issue #8548)
+SELECT
+  70 AS c_0
+FROM
+  (
+    SELECT
+      (
+        EXISTS (
+          SELECT ref_5.c33 AS c_0
+          FROM t5 AS ref_5
+          WHERE (make_timestamp(2001, 7, 13, 17, 53, 31)) = (ref_1.c43)
+        )
+      ) AS c_0
+    FROM
+      (
+        t4 AS ref_0
+        LEFT OUTER JOIN t22 AS ref_1
+          ON (ref_0.vkey = ref_1.vkey)
+      )
+    WHERE
+      (
+        (ref_0.c31) >= (
+          SELECT ref_1.pkey AS c_0
+          FROM t2 AS ref_2
+          WHERE (true) < ((ref_2.c17) ^@ (ref_0.c32))
+          ORDER BY c_0 DESC
+          LIMIT 1
+        )
+      ) IN (
+        SELECT (ref_1.c40) <= (ref_1.c37) AS c_0
+        FROM t7 AS ref_4
+        WHERE NOT ((ref_1.c40) <> (ref_1.c41))
+      )
+  ) AS subq_0
+WHERE
+  (TRUE) < (TRUE);
+ c_0
+---------------------------------------------------------------------
+(0 rows)
+
+DROP TABLE t4;
+DROP TABLE t5;
+DROP TABLE t2;
+DROP TABLE t22;
 DROP TABLE local_table;
 DROP TABLE t0;
 DROP TABLE t1;