Re: slight tweaks to documentation about runtime pruning - Mailing list pgsql-hackers

From Amit Langote
Subject Re: slight tweaks to documentation about runtime pruning
Date
Msg-id CA+HiwqEeb_aD50CWTy2vM_t-WZLG6G=SWCKBuR+cjuA1HaZq7Q@mail.gmail.com
Whole thread Raw
In response to Re: slight tweaks to documentation about runtime pruning  (David Rowley <david.rowley@2ndquadrant.com>)
Responses Re: slight tweaks to documentation about runtime pruning  (Amit Langote <Langote_Amit_f8@lab.ntt.co.jp>)
List pgsql-hackers
On Sun, Dec 9, 2018 at 8:13 PM David Rowley
<david.rowley@2ndquadrant.com> wrote:
> listp1 was scanned twice (loops=2), listp2 was scanned just once.
>
> Now it is true that if the subplan was executed 0 times that it will
> appear as "(never executed)", but do we really need to explain in this
> area that "(never executed)" means loops=0?

Ah, I see what you mean.  So, "(never executed)" is not the only sign
of of run-time pruning occurring.  The value of loops can be different
for different subplans / partitions, and it being lower for a given
subplan means that the subplan has been pruned more number of times.

Thanks,
Amit


pgsql-hackers by date:

Previous
From: David Rowley
Date:
Subject: Re: Should new partitions inherit their tablespace from their parent?
Next
From: Stephen Frost
Date:
Subject: Re: pg_partition_tree crashes for a non-defined relation