Re: pgsql: Fix LATERAL join test in test memoize.sql

David Rowley <drowley@postgresql.org> writes:
> Here we add a simple ORDER BY to stop the planner from being able to
> pullup the lateral subquery.

Hmmm ... we have an agreed syntax for a subquery optimization fence,
and that ain't it.  I wouldn't count on this not breaking again in
the future.  Why not OFFSET 0, which is the usual method?

            regards, tom lane