Fix improper repetition of previous results from a hashed aggregate.

ExecReScanAgg's check for whether it could re-use a previously calculated
hashtable neglected the possibility that the Agg node might reference
PARAM_EXEC Params that are not referenced by its input plan node.  That's
okay if the Params are in upper tlist or qual expressions; but if one
appears in aggregate input expressions, then the hashtable contents need
to be recomputed when the Param's value changes.

To avoid unnecessary performance degradation in the case of a Param that
isn't within an aggregate input, add logic to the planner to determine
which Params are within aggregate inputs.  This requires a new field in
struct Agg, but fortunately we never write plans to disk, so this isn't
an initdb-forcing change.

Per report from Jeevan Chalke.  This has been broken since forever,
so back-patch to all supported branches.

Andrew Gierth, with minor adjustments by me

Report: <CAM2+6=VY8ykfLT5Q8vb9B6EbeBk-NGuLbT6seaQ+Fq4zXvrDcA@mail.gmail.com>

6 files changed, 57 insertions, 5 deletions

diff --git a/src/backend/executor/nodeAgg.c b/src/backend/executor/nodeAgg.c
index 1ec2515090..ce2fc281a4 100644
--- a/src/backend/executor/nodeAgg.c
+++ b/src/backend/executor/nodeAgg.c
@@ -3425,11 +3425,13 @@ ExecReScanAgg(AggState *node)
 			return;
 
 		/*
-		 * If we do have the hash table and the subplan does not have any
-		 * parameter changes, then we can just rescan the existing hash table;
-		 * no need to build it again.
+		 * If we do have the hash table, and the subplan does not have any
+		 * parameter changes, and none of our own parameter changes affect
+		 * input expressions of the aggregated functions, then we can just
+		 * rescan the existing hash table; no need to build it again.
 		 */
-		if (outerPlan->chgParam == NULL)
+		if (outerPlan->chgParam == NULL &&
+			!bms_overlap(node->ss.ps.chgParam, aggnode->aggParams))
 		{
 			ResetTupleHashIterator(node->hashtable, &node->hashiter);
 			return;
diff --git a/src/backend/nodes/copyfuncs.c b/src/backend/nodes/copyfuncs.c
index c7a06442ba..1877fb45e5 100644
--- a/src/backend/nodes/copyfuncs.c
+++ b/src/backend/nodes/copyfuncs.c
@@ -877,6 +877,7 @@ _copyAgg(const Agg *from)
 		COPY_POINTER_FIELD(grpOperators, from->numCols * sizeof(Oid));
 	}
 	COPY_SCALAR_FIELD(numGroups);
+	COPY_BITMAPSET_FIELD(aggParams);
 	COPY_NODE_FIELD(groupingSets);
 	COPY_NODE_FIELD(chain);
 
diff --git a/src/backend/nodes/outfuncs.c b/src/backend/nodes/outfuncs.c
index 50019f4164..29b7712584 100644
--- a/src/backend/nodes/outfuncs.c
+++ b/src/backend/nodes/outfuncs.c
@@ -716,6 +716,7 @@ _outAgg(StringInfo str, const Agg *node)
 		appendStringInfo(str, " %u", node->grpOperators[i]);
 
 	WRITE_LONG_FIELD(numGroups);
+	WRITE_BITMAPSET_FIELD(aggParams);
 	WRITE_NODE_FIELD(groupingSets);
 	WRITE_NODE_FIELD(chain);
 }
diff --git a/src/backend/nodes/readfuncs.c b/src/backend/nodes/readfuncs.c
index c83063e219..6f9a81e3e3 100644
--- a/src/backend/nodes/readfuncs.c
+++ b/src/backend/nodes/readfuncs.c
@@ -2007,6 +2007,7 @@ _readAgg(void)
 	READ_ATTRNUMBER_ARRAY(grpColIdx, local_node->numCols);
 	READ_OID_ARRAY(grpOperators, local_node->numCols);
 	READ_LONG_FIELD(numGroups);
+	READ_BITMAPSET_FIELD(aggParams);
 	READ_NODE_FIELD(groupingSets);
 	READ_NODE_FIELD(chain);
 
diff --git a/src/backend/optimizer/plan/createplan.c b/src/backend/optimizer/plan/createplan.c
index 54d601fc47..47158f6468 100644
--- a/src/backend/optimizer/plan/createplan.c
+++ b/src/backend/optimizer/plan/createplan.c
@@ -5664,6 +5664,7 @@ make_agg(List *tlist, List *qual,
 	node->grpColIdx = grpColIdx;
 	node->grpOperators = grpOperators;
 	node->numGroups = numGroups;
+	node->aggParams = NULL;		/* SS_finalize_plan() will fill this */
 	node->groupingSets = groupingSets;
 	node->chain = chain;
 
diff --git a/src/backend/optimizer/plan/subselect.c b/src/backend/optimizer/plan/subselect.c
index a46cc10820..6edefb1138 100644
--- a/src/backend/optimizer/plan/subselect.c
+++ b/src/backend/optimizer/plan/subselect.c
@@ -82,6 +82,7 @@ static Bitmapset *finalize_plan(PlannerInfo *root,
 			  Bitmapset *valid_params,
 			  Bitmapset *scan_params);
 static bool finalize_primnode(Node *node, finalize_primnode_context *context);
+static bool finalize_agg_primnode(Node *node, finalize_primnode_context *context);
 
 
 /*
@@ -2652,6 +2653,29 @@ finalize_plan(PlannerInfo *root, Plan *plan, Bitmapset *valid_params,
 										 locally_added_param);
 			break;
 
+		case T_Agg:
+			{
+				Agg		   *agg = (Agg *) plan;
+
+				/*
+				 * AGG_HASHED plans need to know which Params are referenced
+				 * in aggregate calls.  Do a separate scan to identify them.
+				 */
+				if (agg->aggstrategy == AGG_HASHED)
+				{
+					finalize_primnode_context aggcontext;
+
+					aggcontext.root = root;
+					aggcontext.paramids = NULL;
+					finalize_agg_primnode((Node *) agg->plan.targetlist,
+										  &aggcontext);
+					finalize_agg_primnode((Node *) agg->plan.qual,
+										  &aggcontext);
+					agg->aggParams = aggcontext.paramids;
+				}
+			}
+			break;
+
 		case T_WindowAgg:
 			finalize_primnode(((WindowAgg *) plan)->startOffset,
 							  &context);
@@ -2660,7 +2684,6 @@ finalize_plan(PlannerInfo *root, Plan *plan, Bitmapset *valid_params,
 			break;
 
 		case T_Hash:
-		case T_Agg:
 		case T_Material:
 		case T_Sort:
 		case T_Unique:
@@ -2812,6 +2835,29 @@ finalize_primnode(Node *node, finalize_primnode_context *context)
 }
 
 /*
+ * finalize_agg_primnode: find all Aggref nodes in the given expression tree,
+ * and add IDs of all PARAM_EXEC params appearing within their aggregated
+ * arguments to the result set.
+ */
+static bool
+finalize_agg_primnode(Node *node, finalize_primnode_context *context)
+{
+	if (node == NULL)
+		return false;
+	if (IsA(node, Aggref))
+	{
+		Aggref	   *agg = (Aggref *) node;
+
+		/* we should not consider the direct arguments, if any */
+		finalize_primnode((Node *) agg->args, context);
+		finalize_primnode((Node *) agg->aggfilter, context);
+		return false;			/* there can't be any Aggrefs below here */
+	}
+	return expression_tree_walker(node, finalize_agg_primnode,
+								  (void *) context);
+}
+
+/*
  * SS_make_initplan_output_param - make a Param for an initPlan's output
  *
  * The plan is expected to return a scalar value of the given type/collation.