Postgres Pro
Downloads
Home   >   mailing lists

Re: Re: [COMMITTERS] pgsql: Rewrite GEQO's gimme_tree function so that it always finds a - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Re: [COMMITTERS] pgsql: Rewrite GEQO's gimme_tree function so that it always finds a
Date
Msg-id 28261.1257782223@sss.pgh.pa.us
Whole thread Raw
In response to Re: [COMMITTERS] pgsql: Rewrite GEQO's gimme_tree function so that it always finds a  (Robert Haas <robertmhaas@gmail.com>)
Responses Re: Re: [COMMITTERS] pgsql: Rewrite GEQO's gimme_tree function so that it always finds a
List pgsql-hackers
Tree view 
Robert Haas <robertmhaas@gmail.com> writes:
> I've stopped the query more than 10 times now and EVERY SINGLE ONE
> finds it in list_concat_unique_ptr().  :-(

Too bad you don't have debug symbols ... it'd be interesting to see
how long that list is.

> It's also using about 12x as much RAM as the GEQO version.

No surprise.  GEQO is designed to constrain its memory use, the
exhaustive planner not so much.
        regards, tom lane

pgsql-hackers by date:

Previous
From: Sonu
Date:
Subject: Re: Specific names for plpgsql variable-resolution control options?
Next
From: Jan Otto
Date:
Subject: Re: drop tablespace error: invalid argument