Fix bug with view locking code.

LockViewRecurese() obtains view relation using heap_open() and passes
it to get_view_query() to get view info. It immediately closes the
relation then uses the returned view info by calling
LockViewRecurse_walker().  Since get_view_query() returns a pointer
within the relcache, the relcache should be kept until
LockViewRecurse_walker() returns. Otherwise the relation could point
to a garbage memory area.

Fix is moving the heap_close() call after LockViewRecurse_walker().

Problem reported by Tom Lane (buildfarm is unhappy, especially prion
since it enables -DRELCACHE_FORCE_RELEASE cpp flag), fix by me.
This commit is contained in:
Tatsuo Ishii 2018-03-31 09:26:43 +09:00
parent 3e256e5506
commit 1b26bd4089
1 changed files with 2 additions and 1 deletions

View File

@ -263,7 +263,6 @@ LockViewRecurse(Oid reloid, Oid root_reloid, LOCKMODE lockmode, bool nowait)
view = heap_open(reloid, NoLock);
viewquery = get_view_query(view);
heap_close(view, NoLock);
context.root_reloid = root_reloid;
context.lockmode = lockmode;
@ -272,6 +271,8 @@ LockViewRecurse(Oid reloid, Oid root_reloid, LOCKMODE lockmode, bool nowait)
context.viewoid = reloid;
LockViewRecurse_walker((Node *) viewquery, &context);
heap_close(view, NoLock);
}
/*