Skip system attributes when applying mvdistinct stats

When estimating number of distinct groups, we failed to ignore system
attributes when matching the group expressions to mvdistinct stats,
causing failures like

  ERROR: negative bitmapset member not allowed

Fix that by simply skipping anything that is not a regular attribute.
Backpatch to PostgreSQL 10, where the extended stats were introduced.

Bug: #16111
Reported-by: Tuomas Leikola
Author: Tomas Vondra
Backpatch-through: 10
Discussion: https://postgr.es/m/16111-687799584c3a7e73@postgresql.org

3 files changed, 30 insertions, 5 deletions

diff --git a/src/backend/utils/adt/selfuncs.c b/src/backend/utils/adt/selfuncs.c
index 7622edb9dd4..126ffad67d0 100644
--- a/src/backend/utils/adt/selfuncs.c
+++ b/src/backend/utils/adt/selfuncs.c
@@ -3931,14 +3931,19 @@ estimate_multivariate_ndistinct(PlannerInfo *root, RelOptInfo *rel,
 	foreach(lc, *varinfos)
 	{
 		GroupVarInfo *varinfo = (GroupVarInfo *) lfirst(lc);
+		AttrNumber	attnum;
 
 		Assert(varinfo->rel == rel);
 
-		if (IsA(varinfo->var, Var))
-		{
-			attnums = bms_add_member(attnums,
-									 ((Var *) varinfo->var)->varattno);
-		}
+		if (!IsA(varinfo->var, Var))
+			continue;
+
+		attnum = ((Var *) varinfo->var)->varattno;
+
+		if (!AttrNumberIsForUserDefinedAttr(attnum))
+			continue;
+
+		attnums = bms_add_member(attnums, attnum);
 	}
 
 	/* look for the ndistinct statistics matching the most vars */
@@ -4018,6 +4023,10 @@ estimate_multivariate_ndistinct(PlannerInfo *root, RelOptInfo *rel,
 			}
 
 			attnum = ((Var *) varinfo->var)->varattno;
+
+			if (!AttrNumberIsForUserDefinedAttr(attnum))
+				continue;
+
 			if (!bms_is_member(attnum, matched))
 				newlist = lappend(newlist, varinfo);
 		}
diff --git a/src/test/regress/expected/stats_ext.out b/src/test/regress/expected/stats_ext.out
index eebf250998e..5b7e1d46608 100644
--- a/src/test/regress/expected/stats_ext.out
+++ b/src/test/regress/expected/stats_ext.out
@@ -217,6 +217,18 @@ SELECT stxkind, stxndistinct
  {d,f}   | {"3, 4": 301, "3, 6": 301, "4, 6": 301, "3, 4, 6": 301}
 (1 row)
 
+-- minor improvement, make sure the ctid does not break the matching
+EXPLAIN (COSTS off)
+SELECT COUNT(*) FROM ndistinct GROUP BY ctid, a, b;
+            QUERY PLAN             
+-----------------------------------
+ GroupAggregate
+   Group Key: ctid, a, b
+   ->  Sort
+         Sort Key: ctid, a, b
+         ->  Seq Scan on ndistinct
+(5 rows)
+
 -- Hash Aggregate, thanks to estimates improved by the statistic
 EXPLAIN (COSTS off)
  SELECT COUNT(*) FROM ndistinct GROUP BY a, b;
diff --git a/src/test/regress/sql/stats_ext.sql b/src/test/regress/sql/stats_ext.sql
index 43ff77c5344..2543dbfffa7 100644
--- a/src/test/regress/sql/stats_ext.sql
+++ b/src/test/regress/sql/stats_ext.sql
@@ -144,6 +144,10 @@ ANALYZE ndistinct;
 SELECT stxkind, stxndistinct
   FROM pg_statistic_ext WHERE stxrelid = 'ndistinct'::regclass;
 
+-- minor improvement, make sure the ctid does not break the matching
+EXPLAIN (COSTS off)
+SELECT COUNT(*) FROM ndistinct GROUP BY ctid, a, b;
+
 -- Hash Aggregate, thanks to estimates improved by the statistic
 EXPLAIN (COSTS off)
  SELECT COUNT(*) FROM ndistinct GROUP BY a, b;