Account for optimized MinMax aggregates during SS_finalize_plan.

We are capable of optimizing MIN() and MAX() aggregates on indexed
columns into subqueries that exploit the index, rather than the normal
thing of scanning the whole table.  When we do this, we replace the
Aggref node(s) with Params referencing subquery outputs.  Such Params
really ought to be included in the per-plan-node extParam/allParam
sets computed by SS_finalize_plan.  However, we've never done so
up to now because of an ancient implementation choice to perform
that substitution during set_plan_references, which runs after
SS_finalize_plan, so that SS_finalize_plan never sees these Params.

The cleanest fix would be to perform a separate tree walk to do
these substitutions before SS_finalize_plan runs.  That seems
unattractive, first because a whole-tree mutation pass is expensive,
and second because we lack infrastructure for visiting expression
subtrees in a Plan tree, so that we'd need a new function knowing
as much as SS_finalize_plan knows about that.  I also considered
swapping the order of SS_finalize_plan and set_plan_references,
but that fell foul of various assumptions that seem tricky to fix.
So the approach adopted here is to teach SS_finalize_plan itself
to check for such Aggrefs.  I refactored things a bit in setrefs.c
to avoid having three copies of the code that does that.

Back-patch of v17 commits d0d44049d and 779ac2c74.  When d0d44049d
went in, there was no evidence that it was fixing a reachable bug,
so I refrained from back-patching.  Now we have such evidence.

Per bug #18465 from Hal Takahara.  Back-patch to all supported
branches.

Discussion: https://postgr.es/m/18465-2fae927718976b22@postgresql.org
Discussion: https://postgr.es/m/2391880.1689025003@sss.pgh.pa.us

diff --git a/src/backend/optimizer/plan/setrefs.c b/src/backend/optimizer/plan/setrefs.c
index 6150bd93f9db35e0aeb04d674af13a1c8e90d417..e65825580dc48dd87989e921b937b0a5252ee69e 100644 (file)
--- a/src/backend/optimizer/plan/setrefs.c
+++ b/src/backend/optimizer/plan/setrefs.c
@@ -1713,22 +1713,14 @@ fix_scan_expr_mutator(Node *node, fix_scan_expr_context *context)
        if (IsA(node, Aggref))
        {
                Aggref     *aggref = (Aggref *) node;
+               Param      *aggparam;
 
                /* See if the Aggref should be replaced by a Param */
-               if (context->root->minmax_aggs != NIL &&
-                       list_length(aggref->args) == 1)
+               aggparam = find_minmax_agg_replacement_param(context->root, aggref);
+               if (aggparam != NULL)
                {
-                       TargetEntry *curTarget = (TargetEntry *) linitial(aggref->args);
-                       ListCell   *lc;
-
-                       foreach(lc, context->root->minmax_aggs)
-                       {
-                               MinMaxAggInfo *mminfo = (MinMaxAggInfo *) lfirst(lc);
-
-                               if (mminfo->aggfnoid == aggref->aggfnoid &&
-                                       equal(mminfo->target, curTarget->expr))
-                                       return (Node *) copyObject(mminfo->param);
-                       }
+                       /* Make a copy of the Param for paranoia's sake */
+                       return (Node *) copyObject(aggparam);
                }
                /* If no match, just fall through to process it normally */
        }
@@ -2623,22 +2615,14 @@ fix_upper_expr_mutator(Node *node, fix_upper_expr_context *context)
        if (IsA(node, Aggref))
        {
                Aggref     *aggref = (Aggref *) node;
+               Param      *aggparam;
 
                /* See if the Aggref should be replaced by a Param */
-               if (context->root->minmax_aggs != NIL &&
-                       list_length(aggref->args) == 1)
+               aggparam = find_minmax_agg_replacement_param(context->root, aggref);
+               if (aggparam != NULL)
                {
-                       TargetEntry *curTarget = (TargetEntry *) linitial(aggref->args);
-                       ListCell   *lc;
-
-                       foreach(lc, context->root->minmax_aggs)
-                       {
-                               MinMaxAggInfo *mminfo = (MinMaxAggInfo *) lfirst(lc);
-
-                               if (mminfo->aggfnoid == aggref->aggfnoid &&
-                                       equal(mminfo->target, curTarget->expr))
-                                       return (Node *) copyObject(mminfo->param);
-                       }
+                       /* Make a copy of the Param for paranoia's sake */
+                       return (Node *) copyObject(aggparam);
                }
                /* If no match, just fall through to process it normally */
        }
@@ -2714,6 +2698,38 @@ set_returning_clause_references(PlannerInfo *root,
 }
 
 
+/*
+ * find_minmax_agg_replacement_param
+ *             If the given Aggref is one that we are optimizing into a subquery
+ *             (cf. planagg.c), then return the Param that should replace it.
+ *             Else return NULL.
+ *
+ * This is exported so that SS_finalize_plan can use it before setrefs.c runs.
+ * Note that it will not find anything until we have built a Plan from a
+ * MinMaxAggPath, as root->minmax_aggs will never be filled otherwise.
+ */
+Param *
+find_minmax_agg_replacement_param(PlannerInfo *root, Aggref *aggref)
+{
+       if (root->minmax_aggs != NIL &&
+               list_length(aggref->args) == 1)
+       {
+               TargetEntry *curTarget = (TargetEntry *) linitial(aggref->args);
+               ListCell   *lc;
+
+               foreach(lc, root->minmax_aggs)
+               {
+                       MinMaxAggInfo *mminfo = (MinMaxAggInfo *) lfirst(lc);
+
+                       if (mminfo->aggfnoid == aggref->aggfnoid &&
+                               equal(mminfo->target, curTarget->expr))
+                               return mminfo->param;
+               }
+       }
+       return NULL;
+}
+
+
 /*****************************************************************************
  *                                     QUERY DEPENDENCY MANAGEMENT
  *****************************************************************************/

diff --git a/src/backend/optimizer/plan/subselect.c b/src/backend/optimizer/plan/subselect.c
index a370b41b0c764845b54ae589216135d215e24872..b279e1b7d717e24679a6eee0a77917b6094cbb5e 100644 (file)
--- a/src/backend/optimizer/plan/subselect.c
+++ b/src/backend/optimizer/plan/subselect.c
@@ -2820,8 +2820,8 @@ finalize_plan(PlannerInfo *root, Plan *plan,
 }
 
 /*
- * finalize_primnode: add IDs of all PARAM_EXEC params appearing in the given
- * expression tree to the result set.
+ * finalize_primnode: add IDs of all PARAM_EXEC params that appear (or will
+ * appear) in the given expression tree to the result set.
  */
 static bool
 finalize_primnode(Node *node, finalize_primnode_context *context)
@@ -2838,7 +2838,26 @@ finalize_primnode(Node *node, finalize_primnode_context *context)
                }
                return false;                   /* no more to do here */
        }
-       if (IsA(node, SubPlan))
+       else if (IsA(node, Aggref))
+       {
+               /*
+                * Check to see if the aggregate will be replaced by a Param
+                * referencing a subquery output during setrefs.c.  If so, we must
+                * account for that Param here.  (For various reasons, it's not
+                * convenient to perform that substitution earlier than setrefs.c, nor
+                * to perform this processing after setrefs.c.  Thus we need a wart
+                * here.)
+                */
+               Aggref     *aggref = (Aggref *) node;
+               Param      *aggparam;
+
+               aggparam = find_minmax_agg_replacement_param(context->root, aggref);
+               if (aggparam != NULL)
+                       context->paramids = bms_add_member(context->paramids,
+                                                                                          aggparam->paramid);
+               /* Fall through to examine the agg's arguments */
+       }
+       else if (IsA(node, SubPlan))
        {
                SubPlan    *subplan = (SubPlan *) node;
                Plan       *plan = planner_subplan_get_plan(context->root, subplan);

diff --git a/src/include/optimizer/planmain.h b/src/include/optimizer/planmain.h
index 80dc4bf4e8833b07de84faba604028d708f7971e..252e3aca559e55cc7f96442ffd3b22a0661ccc5c 100644 (file)
--- a/src/include/optimizer/planmain.h
+++ b/src/include/optimizer/planmain.h
@@ -109,6 +109,8 @@ extern bool innerrel_is_unique(PlannerInfo *root,
  * prototypes for plan/setrefs.c
  */
 extern Plan *set_plan_references(PlannerInfo *root, Plan *plan);
+extern Param *find_minmax_agg_replacement_param(PlannerInfo *root,
+                                                                                               Aggref *aggref);
 extern void record_plan_function_dependency(PlannerInfo *root, Oid funcid);
 extern void record_plan_type_dependency(PlannerInfo *root, Oid typid);
 extern bool extract_query_dependencies_walker(Node *node, PlannerInfo *root);

diff --git a/src/test/regress/expected/aggregates.out b/src/test/regress/expected/aggregates.out
index 462457c9799f2e3480b60e4d371d73f79ffc2d6a..20c217a93b9545fd1f681c6b8986922b4b233cf7 100644 (file)
--- a/src/test/regress/expected/aggregates.out
+++ b/src/test/regress/expected/aggregates.out
@@ -1178,6 +1178,37 @@ NOTICE:  drop cascades to 3 other objects
 DETAIL:  drop cascades to table minmaxtest1
 drop cascades to table minmaxtest2
 drop cascades to table minmaxtest3
+-- DISTINCT can also trigger wrong answers with hash aggregation (bug #18465)
+begin;
+set local enable_sort = off;
+explain (costs off)
+  select f1, (select distinct min(t1.f1) from int4_tbl t1 where t1.f1 = t0.f1)
+  from int4_tbl t0;
+                             QUERY PLAN                              
+---------------------------------------------------------------------
+ Seq Scan on int4_tbl t0
+   SubPlan 2
+     ->  HashAggregate
+           Group Key: $1
+           InitPlan 1 (returns $1)
+             ->  Limit
+                   ->  Seq Scan on int4_tbl t1
+                         Filter: ((f1 IS NOT NULL) AND (f1 = t0.f1))
+           ->  Result
+(9 rows)
+
+select f1, (select distinct min(t1.f1) from int4_tbl t1 where t1.f1 = t0.f1)
+from int4_tbl t0;
+     f1      |     min     
+-------------+-------------
+           0 |           0
+      123456 |      123456
+     -123456 |     -123456
+  2147483647 |  2147483647
+ -2147483647 | -2147483647
+(5 rows)
+
+rollback;
 -- check for correct detection of nested-aggregate errors
 select max(min(unique1)) from tenk1;
 ERROR:  aggregate function calls cannot be nested

diff --git a/src/test/regress/sql/aggregates.sql b/src/test/regress/sql/aggregates.sql
index 0f89b7680e458dc15a91fcd2461e60faecd92186..e870e044248b700f492323e60c63311b63acdaf5 100644 (file)
--- a/src/test/regress/sql/aggregates.sql
+++ b/src/test/regress/sql/aggregates.sql
@@ -394,6 +394,16 @@ select distinct min(f1), max(f1) from minmaxtest;
 
 drop table minmaxtest cascade;
 
+-- DISTINCT can also trigger wrong answers with hash aggregation (bug #18465)
+begin;
+set local enable_sort = off;
+explain (costs off)
+  select f1, (select distinct min(t1.f1) from int4_tbl t1 where t1.f1 = t0.f1)
+  from int4_tbl t0;
+select f1, (select distinct min(t1.f1) from int4_tbl t1 where t1.f1 = t0.f1)
+from int4_tbl t0;
+rollback;
+
 -- check for correct detection of nested-aggregate errors
 select max(min(unique1)) from tenk1;
 select (select max(min(unique1)) from int8_tbl) from tenk1;