Fix the plan-invalidation mechanism to treat regclass constants that refer to

a relation as a reason to invalidate a plan when the relation changes.  This
handles scenarios such as dropping/recreating a sequence that is referenced by
nextval('seq') in a cached plan.  Rather than teach plancache.c all about
digging through plan trees to find regclass Consts, we charge the planner's
setrefs.c with making a list of the relation OIDs on which each plan depends.
That way the list can be built cheaply during a plan tree traversal that has
to happen anyway.  Per bug #3662 and subsequent discussion.

2 files changed, 31 insertions, 0 deletions

diff --git a/src/test/regress/expected/plancache.out b/src/test/regress/expected/plancache.out
index a96ba2a5da..d7d7be9252 100644
--- a/src/test/regress/expected/plancache.out
+++ b/src/test/regress/expected/plancache.out
@@ -202,3 +202,20 @@ drop schema s1 cascade;
 NOTICE:  drop cascades to table s1.abc
 drop schema s2 cascade;
 NOTICE:  drop cascades to table abc
+-- Check that invalidation deals with regclass constants
+create temp sequence seq;
+prepare p2 as select nextval('seq');
+execute p2;
+ nextval 
+---------
+       1
+(1 row)
+
+drop sequence seq;
+create temp sequence seq;
+execute p2;
+ nextval 
+---------
+       1
+(1 row)
+
diff --git a/src/test/regress/sql/plancache.sql b/src/test/regress/sql/plancache.sql
index e285c071f6..fc57279d98 100644
--- a/src/test/regress/sql/plancache.sql
+++ b/src/test/regress/sql/plancache.sql
@@ -123,3 +123,17 @@ execute p1;
 
 drop schema s1 cascade;
 drop schema s2 cascade;
+
+-- Check that invalidation deals with regclass constants
+
+create temp sequence seq;
+
+prepare p2 as select nextval('seq');
+
+execute p2;
+
+drop sequence seq;
+
+create temp sequence seq;
+
+execute p2;