Amit Langote <Langote_Amit_f8@lab.ntt.co.jp> writes:
> Currently find_inheritance_children() is smart enough to skip a child
> table that it finds has been dropped concurrently after it gets a lock on
> the same. It does so by looking up the child relid in syscache. It seems
> it should also check if the table is still in the list of children of the
> parent. Doing so by scanning the pg_inherits(inhparent) index may likely
> be inefficient. So, how about adding that syscache on
> pg_inherits(inherelid, inhparent) [1]?
I doubt that a syscache would fix the performance issue there; it wouldn't
get referenced enough to be likely to have the desired tuple in cache.
I wonder whether we could improve matters by rechecking validity of the
pg_inherits tuple (which we saw already and could presumably retain the
TID of). There is at least one place where we do something like that now,
IIRC.
regards, tom lane
