On Thu, 22 Jul 2021 at 01:54, David Rowley <dgrowle...@gmail.com> wrote: > Maybe the planner would need to be involved in making the decision of > if the bitmap index scan should tuck away a carbon copy of the > resulting TIDBitmap after the first scan. That way on rescan we could > just make a copy of the cached version and return that. That saves > having to modify the callers to tell them not to damage the returned > TIDBitmap.
Oh but, meh. Caching could blow out work_mem... We might end up using work_mem * 2. David
