Paul Guo <paulguo(at)gmail(dot)com> writes:
> Postgres seems to always optimize ORDER BY + LIMIT as top-k sort.
> Recently I happened to notice
> that in this scenario the output tuple number of the sort node is not
> the same as the LIMIT tuple number.
No, it isn't, and your proposed patch is completely misguided.
The cost and rowcount estimates for a plan node are always written
on the assumption that the node is run to completion.
regards, tom lane