Re: performance regression in 9.2 CTE with SRF function

Pavel Stehule <pavel.stehule@gmail.com> writes:
>> no, there is strange estimation
>> 
>> SubPlan 2
>> ->  CTE Scan on pl pl_1  (cost=0.00..468.59
>> rows=89000 width=4) (actual time=0.023..8.379 rows=566 loops=1000)
>> Output: foo(pl_1.a)

Nothing strange about it.  89 rows out from the underlying CTE (which
appears to be dead accurate) times 1000 for the SRF's expansion.

> respective why estimation is unstrable

It isn't --- you are looking at two different subplans there, one with
a SRF in it and one without.
        regards, tom lane