Re: Adding pipelining support to set returning functions - Mailing list pgsql-hackers

From Martijn van Oosterhout
Subject Re: Adding pipelining support to set returning functions
Date
Msg-id 20080411111122.GD23648@svana.org
Whole thread Raw
In response to Re: Adding pipelining support to set returning functions  (Hannu Krosing <hannu@krosing.net>)
List pgsql-hackers
On Fri, Apr 11, 2008 at 01:00:04PM +0300, Hannu Krosing wrote:
> > i asked myself basically the same question some time ago.
> > pipelining seems fairly impossible unless we ban joins on those
> > "plugins" completely.
>
> Not really, just they have to be "materialized" before joins, or
> streaming node has to be at the driving side of the join, so you can
> fetch one tuple and then join it to index or hash lookup

I thought these was code in the planner already that said: if node A
requires seeking of subnode B and B doesn't support that, insert
Materialize node. Naturally there's a cost to that, so it would favour
plans that did not require seeking...

Have a nice day,
--
Martijn van Oosterhout   <kleptog@svana.org>   http://svana.org/kleptog/
> Please line up in a tree and maintain the heap invariant while
> boarding. Thank you for flying nlogn airlines.

pgsql-hackers by date:

Previous
From: "Zeugswetter Andreas OSB SD"
Date:
Subject: Re: Index AM change proposals, redux
Next
From: Magnus Hagander
Date:
Subject: stat() vs cygwin