Received: from maia.hub.org (maia-3.hub.org [200.46.204.243])
	by mail.postgresql.org (Postfix) with ESMTP id E7EFDB5DE2E
	for <pgsql-performance-postgresql.org@mail.postgresql.org>;
	Thu,  1 Sep 2011 20:14:06 -0300 (ADT)
Received: from mail.postgresql.org ([200.46.204.86])
	by maia.hub.org (mx1.hub.org [200.46.204.243]) (amavisd-maia,
	port 10024) with ESMTP id 94608-02-2
	for <pgsql-performance-postgresql.org@mail.postgresql.org>;
	Thu,  1 Sep 2011 23:14:00 +0000 (UTC)
X-Greylist: from auto-whitelisted by SQLgrey-1.7.6
Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130])
	by mail.postgresql.org (Postfix) with ESMTP id 69796B5DCE9
	for <pgsql-performance@postgresql.org>;
	Thu,  1 Sep 2011 20:14:00 -0300 (ADT)
Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1])
	by sss.pgh.pa.us (8.14.2/8.14.2) with ESMTP id p81NDuph026941;
	Thu, 1 Sep 2011 19:13:56 -0400 (EDT)
To: Mark Kirkwood <mark.kirkwood@catalyst.net.nz>
cc: PGSQL Performance <pgsql-performance@postgresql.org>
Subject: Re: 8.4 optimization regression? 
In-reply-to: <5156.1314829311@sss.pgh.pa.us> 
References: <4E5331FF.3000900@catalyst.net.nz>
 <15708.1314155740@sss.pgh.pa.us>
	<4E5DCFA4.5090806@catalyst.net.nz> <5156.1314829311@sss.pgh.pa.us>
Comments: In-reply-to Tom Lane <tgl@sss.pgh.pa.us>
	message dated "Wed, 31 Aug 2011 18:21:51 -0400"
Date: Thu, 01 Sep 2011 19:13:56 -0400
Message-ID: <26940.1314918836@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>
X-Virus-Scanned: Maia Mailguard 1.0.1
X-Spam-Status: No, hits=-2.404 tagged_above=-10 required=5
	tests=BAYES_00=-1.9, RP_MATCHES_RCVD=-0.504
X-Spam-Level: 
X-Archive-Number: 201109/2
X-Sequence-Number: 44778

I wrote:
> Mark Kirkwood <mark.kirkwood@catalyst.net.nz> writes:
>> [ assorted examples showing that commit
>> 7f3eba30c9d622d1981b1368f2d79ba0999cdff2 has got problems ]

> ...
> So, not only are you correct that we should revert the changes to
> eqjoinsel_inner, but what's happening in eqjoinsel_semi is wrong too.

I've retested these examples with the patches I committed yesterday.
Six of the eight examples are estimated pretty nearly dead on, while the
other two are estimated about 50% too high (still a lot better than
before).  AFAICT there's no easy way to improve those estimates further;
eqjoinsel_semi just plain hasn't got enough information to know how many
matches there will be.

			regards, tom lane